The Exploitation of the Labour of Love

Vasti Calitz

Thesis presented in partial fulfillment of the requirements for the degree of Master of Philosophy in the Faculty of the Humanities at Stellenbosch University.

Supervisor: Prof Louise Du Toit

March 2018 Stellenbosch University https://scholar.sun.ac.za

Declaration

By submitting this thesis electronically, I declare that the entirety of the work contained therein is my own, original work, that I am the sole author thereof (save to the extent explicitly otherwise stated), that reproduction and publication thereof by Stellenbosch University will not infringe any third party rights and that I have not previously in its entirety or in part submitted it for obtaining any qualification.

Abstract

This thesis seeks to establish the wrongfulness of an unequal division of nurturing work between members of heterosexual couples. Nurturing work is the overlapping constellation of housework, care, and emotion work, each of which women do more of than their male partners. I turn to feminist political philosophy (specifically Susan Moller Okin) to show that justice requires, at minimum, that the vulnerability women experience as a result of marriage needs to be mitigated by the state, and that the equal distribution of nurturing work needs to be facilitated by labour law. However, this is not enough to establish whether or not one wrongs one’s partner by allowing her to do more nurturing work. In order to prove this, I rely on Ruth Sample’s work to show that an unequal division of labour constitutes degradation of women in three ways. Firstly, it constitutes taking advantage of an existing injustice by gaining the benefit of receiving more care than one gives because one’s female partner was socialised into giving it. Secondly, an inequality of nurturing work is also an inequality in status accord, and if such inequality is gendered, it confirms for oneself and one’s partner, as well as other witnesses, the relative lesser importance (and therefore inferiority) of women. This is also degradation. Thirdly, I argue, using Miranda Fricker and Sandra Bartky, that a gendered distribution of nurturing work contributes to the hermeneutical marginalisation of women, which also constitutes a degradation of women. I thus prove a strong moral obligation to refrain from degrading one’s partner, and therefore a strong moral obligation to not allow one to be taken care of more than one takes care of one’s partner. In the last chapter I show that nurturing work is significant for improving the quality of a relationship, as well as for contributing to one’s human flourishing. I argue this because even if the background conditions are not such that an unequal division of nurturing work would be degradation, there are very good reasons to become good at nurturing work, since it contributes to the flourishing of the individual as well as the relationship.

Stellenbosch University https://scholar.sun.ac.za

Opsomming

Hierdie tesis poog om te wys dat daar ‘n ongelyke verdeling van koesterende werk is tussen mense in heteroseksuele verhoudings en dat dié ongelyke verdeling onregmatig is. Koesterende werk verwys na ‘n oorvleuelende konstellasie van huiswerk, sorg en emosie-werk. Die tesis argumenteer dat vrouens meer van elkeen van dié werke doen as mans. Ek steun op feministiese politieke filosofie (veral die van Susan Moller Okin) om te wys dat, in terme van basiese geregtigheid, die staat verplig is om die kwesbaarheid te versag wat vrouens ervaar as ‘n gevolg van die huwelik, en verder dat arbeidsreg die gelyke verdeling van koesterende werk moet bevorder. Ek wys verder dat die reg onvoldoende is om te bevestig of jy jou gade skade aandoen deur haar toe te laat om meer koesterende werk as jy te doen. Ek steun in dié verband op die werk van Ruth Sample, aangesien Sample aantoon dat vrouens op drie maniere onderdruk word deur die ongelyke verdeling van koesterende werk. Eerstens, omdat vrouens gesosialiseer is om meer koesterende werk as mans te doen, is die ongelyke verdeling van koesterende werk die verdieping van ‘n bestaande ongeregtigheid. Die ongelyke verdeling van koesterende werk kom tweedens neer op ‘n ongelyke verdeling van status. Die ongelyke verdeling van koesterende werk bevestig as sulks dat vrouens minder belangrik (en dus minderwaardig) is as mans. Dit is ook onderdrukking. Ek argumenteer derdens, met verwysing na Miranda Fricker en Sandra Bartky, dat die ongelyke verdeling van koesterende werk bydra tot die hermeneutiese marginalisering van vrouens, en dat dit ook bydra tot die onderdrukking van vrouens oor die algemeen. Ek ontbloot dus ‘n sterk morele plig om vrouens nie te onderdruk nie, en dan ook ‘n sterk morele plig om nie toe te laat dat mens meer gekoester word deur vrouens as wat mens vrouens koester nie. In die finale hoofstuk dui ek aan dat koesterende werk ‘n wesenlike bydrae kan maak tot die kwaliteit van ‘n verhouding, en ook tot die kwaliteit van ‘n mens se lewe. Dit is belangrik, want selfs as die ongelyke verdeling van koesterende werk nie tot vrouens se onderdrukking gelei het nie, sou daar steeds baie goeie redes wees om goed te word in koesterende werk, aangesien dit bydra tot die ontwikkeling van die individu asook die verhouding.

Stellenbosch University https://scholar.sun.ac.za

TABLE OF CONTENTS Introduction ...... 1 Chapter One: Groundwork ...... 5 Introduction ...... 5 1.1. Defining Nurturing Work ...... 6 1.2. Empirical evidence of gendered distribution of labour ...... 11 1.3. The biological objection ...... 21 Conclusion ...... 23 Chapter Two: Contract ...... 25 Introduction ...... 25 2.1. Marriage and contract ...... 26 2.2. The social contract and the family ...... 39 Conclusion ...... 50 Chapter Three: Exploitation of Nurturing Work ...... 53 Introduction ...... 53 3.1. Economic exploitation of emotion work ...... 54 3.2. Alternative accounts of exploitation ...... 63 3.3. Exploitation as contributing to the oppression of women ...... 70 Conclusion ...... 74 Chapter Four: The Ethics of Care ...... 77 Introduction ...... 77 4.1. Money can’t buy you nurturing work ...... 77 4.2. The ethics of care ...... 82 4.3. A weak obligation to perform nurturing work ...... 91 Conclusion ...... 92 Conclusion ...... 94 Reference List ...... 100 Stellenbosch University https://scholar.sun.ac.za

INTRODUCTION

I first became aware of the notion of emotion work on an Internet forum where women discussed the kinds of work they did in their romantic relationships that were not reciprocated.1’Emotion work is the work put in to elevate the status of others as well as to engender positive emotions and soothe negative emotions. These women were describing effort they put into their relationships that was exhausting, time- consuming, essential to the functioning of a household and a partnership, hardly ever noticed, and rarely reciprocated by their partners. This was revelatory to me because it exposed the functioning of relationships of those around me, as well as my own. I realised both that I was doing such unacknowledged work in certain relationships, and not acknowledging such work in other relationships. It was also clear that it was gendered. Not only do women do more housework than men (a fact of which many of us are aware), but they also do more emotion work. When I became aware of the unequal distribution of such nurturing work2, it was immediately clear to me that those of my relatives who are a generation older than I exhibited such a distribution of labour. This was not wholly surprising to me, since these family members of mine are not necessarily feminists. However, it soon became apparent to me that this was not a problem that had stopped with the previous generation – my female friends in their mid-twenties, who were in serious relationships with feminist men experienced a similar inequality in the division of labour in their relationships. It was strange to me that men and women who did not believe in gender roles or in relevant innate differences in abilities between men and women still so clearly exhibited a gendered division of labour. It seemed obviously unfair to me that one partner could do most of the ‘taking care of’, and that the other partner received most of the benefit of being ‘taken care of’, and I wanted to determine philosophically why it was wrong. Some months before writing this, I met a couple that caused me to refine my question. The couple was in a 24/7 master-slave relationship, with the male partner in

1 This weblog can be found at https://drive.google.com/file/d/0B0UUYL6kaNeBTDBRbkJkeUtabEk/view?pref=2&pli=1

2 This is a term I will unpack at length in Chapter 1, but briefly, it is a constellation of three overlapping kinds of work: housework, care, and nurturing work.

1 Stellenbosch University https://scholar.sun.ac.za the master role and the female partner in the slave role. Let’s call them Tom and Emma. Before meeting them, I naturally assumed that the balance of power would be in Tom’s favour, and that the relationship would be exploitative of Emma in some way. However, as soon as I met Emma, I knew it wasn’t the case. She was a vivacious, extraverted, and vibrant woman wearing bright red lipstick and a tight dress. Tom, on the other hand, was a poorly dressed, balding twenty-five-year-old who was shy and bland. He preferred to listen to Emma speak rather than speaking himself, even as Emma requested that we ignore her and speak only to him. She told us the rules that dictated her relationship, and these included putting her master in charge of what she ate (she was only allowed sugar on Fridays) and the household budget (she handed over her paycheck to him as soon as she got it). He was not to leave her alone in the flat because she needed company at all times. Interestingly, this couple was polyamorous, so even though Tom putatively had this great power over Emma, he did not even assert the power that many men (and women) do in romantic relationships to prevent her from having sexual relationships with other people. Of course, there were some sexual elements to the rules of their relationship, but most of the rules seemed to be about Emma handing over the responsibility for taking care of herself to Tom – in other words, she was using their master/slave structure as a way of extracting extra care from her male partner that she would not usually get. I learned later from acquaintances that she had attempted this in previous relationships, but Tom was the first who acquiesced to the relationship structure, and that she had sought it out because she suffers from depression and wanted an escape from the responsibility of caring for herself at times. This was the first couple I had met where the distribution of nurturing work was skewed the other way, where the male partner was doing much more of this type of work than the female partner. In the cases of my friends and their feminist partners, their unequal distribution of labour seemed obviously wrong, but with this couple it was not clear to me that it was wrong for Tom to be doing more work than Emma. For this reason, I will use this thesis to determine whether an unequal division of nurturing work is wrong as such, or if it is wrong because it is an instance of patriarchal power over women. In the process of writing, I have vacillated between these two positions, but this thesis will conclude that whereas the wrongfulness of an

2 Stellenbosch University https://scholar.sun.ac.za unequal division of labour because of gender is clear, it is less clear whether it is wrong as such. This thesis is structured as follows: in the first chapter I explicate the necessary conceptual and empirical groundwork for the theoretical work that follows in later chapters. I provide a definition of nurturing work, which is a term I have coined to refer to three overlapping kinds of work, namely, housework, emotion work, and care. I also provide an overview of sociological data which proves that the division of nurturing work is gendered, and so show that women on average do more emotion work, more housework, and more care work than their male partners. The purpose of this is to emphasise the relevance and indeed urgency of this philosophical work. The rest of the thesis attempts to establish whether an unequal distribution of nurturing work is wrongful, either because it is gendered, or wrong as such. My strategy in answering this question has been to test my intuitive responses as to why it is wrong: because it is unjust, because it is exploitative, and because nurturing work is the kind of work that ought to be reciprocated. The second chapter tests the intuition that the unequal distribution of nurturing work is unjust by turning to liberal feminism. I look at two feminist liberals who both use the liberal device of the contract in order to determine how justice applies to intimate relationships, and thereby I determine if the wrongfulness of a gendered distribution of labour is rooted in its unjustness. The first feminist theorist I look at is Jean Hampton, who uses a ‘contractarian test’ to determine if the requirements of justice are met by one’s intimate relationship. I then turn to Susan Moller Okin, who uses the social contract to prove the existence of certain gendered injustice in the institution of the family, particularly in the division of labour. I conclude that Hampton’s contractarian test is not suited to test the fairness of the division of nurturing work, but that Okin’s political philosophy shows that there is an obligation on behalf of the state to facilitate a more equal division of nurturing work via labour law, as well as to mitigate the vulnerability of women caused by marriage. However, political philosophy is ill-suited to determine the wrongfulness of a division of nurturing work in an interpersonal relationship. For this reason, in Chapter Three, I look at whether or not such unequal division of labour can be considered exploitative, which would establish the wrongfulness on the interpersonal level of the relationship. To this end, I utilise

3 Stellenbosch University https://scholar.sun.ac.za thinkers in materialist feminism and their analyses of the modes of production in the home, and how Marx’s theory of exploitation can be extended to intimate relationships. Next, I look at alternative accounts of exploitation to determine if, firstly, the unequal distribution of labour in couples counts as exploitation under these accounts, and if they can successfully explain its wrongfulness. In doing so, I show how the unequal division of nurturing work contributes to the oppression of women by contributing to their hermeneutical marginalisation. I conclude in this chapter that the gendered division of nurturing work is wrong because it constitutes exploitation as well as degradation, and contributes to the oppression of women. The fourth chapter turns to the particular nature of nurturing work in order to determine if there is something about it that means that, firstly, it is non-distributable, and secondly, that it is a moral activity. Using the ethics of care, I explore whether its nature as a moral activity imposes particular moral obligations to provide such work for people with whom one is intimate. I will show how insights from the ethics of care can at best impose a weak obligation to do nurturing work for one’s partner, but that being good at nurturing is important for the flourishing of the individual and of the relationship. There may well be other grounds for arguing for a strong obligation to provide nurturing work for one’s intimates, or at least to reciprocate such work when one is able to do so. The purpose of this paper is to test what seemed to be the most promising existing feminist philosophy to justify such an obligation, and doing so has yielded a strong moral obligation to refrain from distributing nurturing work in a way that is gendered because it is exploitative and contributes to the oppression of women, and a weak general moral obligation to provide nurturing work for one’s partner. Having covered these existing fields of feminist philosophy, I contend that this paper allows for further work that could enrich our understanding of our moral obligations to provide nurturing work for our intimates.

4 Stellenbosch University https://scholar.sun.ac.za

CHAPTER ONE: GROUNDWORK

INTRODUCTION

The aim of this chapter is to lay down the necessary empirical groundwork for the theoretical and normative work that follows in the subsequent chapters. This will entail explaining some of the key terms I will be relying on, specifically nurturing work, which is a term I have coined to refer to an interlinked constellation of emotion work, housework, and care, each of which will be explained. This chapter aims to show that it is still the status quo that in monogamous heterosexual relationships, female partners do far more nurturing work than their male partners. In order to show this, I will be relying on sociological studies from the last two decades done on the kinds of work, and their relative quantities, done by partners in heterosexual relationships. These studies will cover each of the elements of nurturing work, and show that one’s gender is the greatest predictor of how much nurturing work one does, rather than relative income, relative time constraints, and other factors that it could be ascribed to. The scope of this thesis is limited to cohabitating heterosexual couples, married or unmarried. This is possible since, though cohabiting couples sometimes have more egalitarian ideologies than married couples, their distribution of labour remains equally traditional (Miller and Sassler, 2016:696). This chapter will explore the inequality of the distribution of different kinds of work, including housework (including, but not limited to, cleaning, laundry, cooking, shopping, and repairs), childcare, and nurturing work. I use the words ‘wife’, ‘husband’, and ‘partner’ at various points in this paper, but I intend them to refer to the same thing: cohabiting long-term heterosexual partners who are not necessarily married. The words ‘wife’ and ‘husband’ usually refer to married people, but since unmarried couples adopt similar gender roles as married couples, these words need not be exclusively attached to marriage for the purpose of this paper. In limiting the scope of my thesis as such, I have excluded a number of other kinds of romantic relationships, including same-sex couples and multiple-partner polyamorous couples. The reason for excluding same-sex couples is that the unequal distribution of housework and nurturing work interestingly simply does not apply to same-sex couples (Kurdek, 2005:252), which supports the angle that this paper is

5 Stellenbosch University https://scholar.sun.ac.za taking: this is a gendered issue, and is a case of gendered injustice. Further, there simply is not sufficient research about polyamorous multiple-partner relationships to use as a comparison. The findings about heterosexual, monogamous couples can therefore not be extrapolated to these other types of long-term relationships.

1.1. DEFINING NURTURING WORK

I have coined this term to refer collectively to three kinds of work performed in couples that share a household. The three kinds of work are housework, care, and emotion work, and I am treating them collectively rather than separately because of the ways in which they overlap and are intertwined with each other. I will explain each kind of work below, and explain where the overlaps reside, starting with the most complex: emotion work.

1.1.1. EMOTION WORK

Emotion work is a term first used by Arlie Hochschild, a sociologist, who used it to refer to the management of emotions by the self, and also the work that is done to engender and manage emotions in others. This is a completely different way of thinking about how emotions operate: we typically think of them as spontaneous and largely outside of our control. However, Hochschild argues that emotions are often not spontaneous at all, but are rather subject to ‘feeling rules’, with greater or lesser success (1979:564). This is exemplified by someone “trying to like” someone, or “trying not to feel sad”, or attempting to get into an appropriate mood for a party. We are also reminded of these rules when we tell ourselves or others, “You have no right to be jealous”, or “Your anger is irrational” (Hochschild, 1979:564). Such expressions indicate that we recognise feeling rules and attempt to adjust our emotions accordingly. We do this in social interactions with others and even privately. These are all kinds of emotion work we do on ourselves. A further component of emotion work is the work done on other people, which is what is of primary interest for this thesis. These are the ‘activities that are concerned with the enhancement of others’ emotional well-being and with the provision of emotional support’ (Erickson, 2005:338). Also known as ‘interpersonal emotion management’, it is an activity that requires time, effort, and skill (Erickson,

6 Stellenbosch University https://scholar.sun.ac.za

2005:338). This includes comforting, encouraging, showing interest, affirming, showing affection, and empathising. This is not entirely distinct from work done on the self: emotion work done on the self is a form of ‘deep acting’, which means that it is not simply a matter of displaying the correct emotion (such as smiling), but actually feeling the correct emotions (Hochschild, 1979:569). Empathising would, for example, involve actually aligning one’s actual emotions with the struggles of others, rather than only expressing the appropriate words and gestures. For Hochschild, emotion work is politicised rather than neutral, and tied up with the distribution of power in four ways: firstly, it is gendered, where women are expected to take up more of the kind of emotion work that engenders positive emotions and elevated status (which Hochschild refers to as ‘caring’) (2003b:163). According to a study by Erickson, women are held accountable for emotion work (both towards their own and others’ emotions) in a way men are not (2005:339). Secondly, women learn how to use emotion work to compensate for the fact that they have fewer material resources, where they offer their emotion work as a gift in exchange for material resources (Hochschild, 2003b:163). Thirdly, women are vulnerable to the displacement of other people’s emotions, where they are forced to be the recipient of complaint, or anger, or neediness, whereas men experience a ‘status shield’ that protects them from the negative emotions of others (Hochschild, 2003b:181). Finally, the power differential between men and women means that the nurturing work that women do is far more likely to be invisible than when men do the same work. A characteristic of emotion work is that it must appear effortless in order to retain its value. Emotion work that comes across as effortful seems insincere. It must appear to emanate spontaneously from within in order to be effective – for example, as ‘reflections of interpersonal relationships or love’ (Erickson, 2005:349). Put differently, the receiver of emotion work wants to experience the interest shown as flowing from a sincere and deeply felt solidarity between the parties. There is some evidence that men do see it as such – they see emotion work as a feature of interpersonal expression that is not instrumental. However, women are far more likely to conceptualise emotion work as work, probably because they are held accountable for its performance in ways that men are not (Erickson, 2005:339). Further, women do significantly more emotion work than men do, perhaps because it is seen as ‘work’ – in other words, not contingent on leisure, inclination or convenience. Some see the

7 Stellenbosch University https://scholar.sun.ac.za labeling of this activity as ‘work’ as degrading the activity (probably because work is seen as something one is compelled to do, not necessarily something that one wants to do), but the more satisfied two people in a relationship, are the more likely it is that they are doing a great deal of this emotion work (Erickson, 2005:338). As Erickson argues, emotional support should be reconceptualised as work: it ‘does not emanate from within, but must be managed, focused and directed so as to have the intended effect on the recipient’ (Erickson, 2005:349). It takes time, effort, and skill. Hochschild distinguishes between ‘emotion work’ and ‘emotional labour’, where the former takes place in personal relationships, and the latter is commodified, particularly in the service industry. Her work on the commodification of emotional labour will be discussed in detail in the third chapter of this thesis, where this very distinction will be challenged. For the purposes of this thesis, I will entirely exclude the kinds of emotion work that are intended to engender negative feelings or status deflation (such as inducing guilt as a form of punishment in a child). Rather, when I use the term ‘emotion work’ in this thesis, it has the following characteristics (as summarised from above): 1) it is targeted at the interlocutor’s emotions, specifically to induce a positive emotion, 2) it is done with an attitude of willingness; and 3) it must be effective in bringing about those emotions, or at least, its effectiveness must of be of concern to the one doing emotion work. The positive emotions that can be induced include feeling cared for, listened to, appreciated, loved, or comforted, amongst other emotions. The one doing the emotion work must also manage her own emotions so that she feels the appropriate feelings when doing the work: she must, for example, suppress impatience, distraction, and resentment, and cultivate concentrated attention, affection, and willingness. Finally, it is not enough to intend to produce a certain result, she must actually do so. This final element I have borrowed from the ethics of care, which sees care as relational, which means that it must be effective (Tronto, 1993:108). This means that it cannot be the ‘thought that counts’ – it actually has to bring about the desired emotion. It also means that the one doing the emotion work must be receptive to how her work is received, and must alter her strategy to make it more effective if necessary. Note that I have not specified what kinds of activities constitute emotion work. Some unambiguous examples would include the following: noticing someone is out of sorts, asking them what is wrong, helping them determine the cause of their

8 Stellenbosch University https://scholar.sun.ac.za negative emotion, and doing work to eradicate the negative emotion. For example, drawing someone out about their frustration about their boss, which would entail being indignant with one’s partner about their boss. Another example might be boosting someone’s ego if they have lost a job or received criticism, so that one elevates the other’s status to counteract the status deflation that occurred as a result of that event, through, for example, shared indignation, and expressing one’s own indignation on behalf of the other, in one’s own words. However, any kind of activity, if done in such a way that has those three characteristics mentioned above could constitute emotion work. I will point out later how this means that emotion work can overlap with other kinds of nurturing work.

1.1.2. CARE

I have decided to treat care as somewhat distinct from emotion work, since care is often associated with the needy body. Emotion work can certainly be expected to enhance one’s life, but it need not be linked to needs. I am thus using care to refer to ways in which one takes responsibility for someone else’s bodily needs. This would certainly include taking care of someone when they are ill, such as administering to them in their sick bed, or by going out to buy medication they may need. It may also involve taking extra responsibility for a partner if they suffer from mental illness and are sometimes less able to take responsibility for themselves. Caring is thus also especially relevant when the body is more needy than usual, for example, with the very young, the very old, the sick and the vulnerable. It would also include taking responsibility for maintenance of someone else’s health, by, for example, ensuring that they take their chronic medication, or eat healthily, or exercise. I will show in detail how this particular activity of taking responsibility for another’s health is gendered in section 1.2.3. Caring can overlap with emotion work, but need not necessarily. For instance, a caregiver may lay her hand on her partner’s fevered brow both as a gesture of sympathy and to assess the kind of caregiving she needs to do. If caring for an ill partner is offered as a gift, as an intentional way to relieve the other’s burden, and if it were received as such, then it would also be emotion work. However, it is trite to say that this kind of caring work often does not engender positive emotions and rather provokes irritation – the response to a remonstration to not have a second helping of

9 Stellenbosch University https://scholar.sun.ac.za red meat or to add salt, or have another whiskey, is probably never a warm feeling of gratitude expressed by a loving ‘thank you’. The purpose of this kind of remonstration is also clearly not to engender a positive emotion, but rather to ensure the health and longevity of one’s partner.

1.1.3. HOUSEWORK

Housework is arguably the most time-consuming of all the kinds of work, but it is also distinct in that it need not be other-centered. Rather than being work that one does explicitly for one’s partner and for the sake of the relationship, housework is a collection of tasks and activities that need to be done, and can be divided up between partners. I define housework as any work that goes into creating, maintaining, and improving the home (as opposed to the house). It is about the cultivation of a certain kind of dwelling-space, and involves the often very practical work that makes it possible. This would, of course, include cleaning, laundry, repairs, gardening, making a space more beautiful, and maintenance of the building, such as repainting the walls. It involves the mental work of taking stock of what needs to be done, and anticipating needs such as making sure a person has the clothes they need for work washed and ironed, and making sure cleaning supplies are in stock. Cooking is an interesting and complicated activity that at least resides somewhere in between housework and caring, and can overlap with emotion work as well. It can be done as a form of caring – for example, if one’s partner has particular dietary requirements that one accommodates in one’s cooking for them, such as weighing out portions or cooking very healthy food. Taking account of others’ physical needs also takes place in the shopping one does for the household and catering to preferences in doing so. Cooking is also interesting because it is one of the largest tasks that needs to be done for household, since it occurs once or more times a day and can be time-consuming. However, it is also one of the more enjoyable household activities, and for that reason is sometimes more equally distributed than less pleasant tasks like cleaning. Cooking can also constitute emotion work if one cooks one’s partner’s favourite meal in express appreciation for something they did, or to make them feel special. Housework can constitute emotion work when it is offered as a gift. For example, in the event that housework is evenly distributed between partners, and one

10 Stellenbosch University https://scholar.sun.ac.za partner takes on some of the other’s responsibilities because that partner is particularly stressed or depressed, then it is done in part to engender positive emotions for one’s partner. An example might be where if the female partner has the task of driving her children around after school, but has a particularly stressful week in her job, then her male partner taking over that task for the week is both a practical decision as well as an instance of emotion work targeted at her making feel cared for and the importance of her work acknowledged. The various kinds of nurturing work therefore sometimes fall into only one of these categories, but often overlap, and the intention and effect of the activity can mean that all three categories overlap. I will now present empirical data to show how each of these categories, and therefore nurturing work as a whole, is a gendered activity.

1.2. EMPIRICAL EVIDENCE OF GENDERED DISTRIBUTION OF LABOUR

1.2.1. HOUSEWORK

The easiest distribution of labour to measure is that of housework, and as such there is the most data on this subject. The most effective kind of study to measure housework is a time-diary study, where each partner has to account for every hour of their day, since both men and women over-report how much housework they do. One might expect that the distribution of household labour is vastly unequal between working men and their stay-at-home wives, but research also shows that there is a vast discrepancy in hours spent on unpaid work between partners who both work full-time. When a couple lists their responsibilities of household labour, their lists usually have a more or less equal number of items (Kimmel, 2000:128). However, the items on the male partner’s list are usually tasks that occur once a week or once every few weeks – for example, cleaning the pool, mowing the lawn, changing the oil in the car, or repairing broken appliances. In contrast, the items on the female partner’s list are usually daily activities, such as cooking, shopping, and cleaning. Cooking involves one to three meals a day, cleaning occurs every day and shopping often occurs multiple times a week. That means that, despite the number of items being equivalent, the number of hours that go into the respective lists differ vastly.

11 Stellenbosch University https://scholar.sun.ac.za

Men often frame their role in the household as “helping out” or “pitching in” – this implies that the work is actually the woman’s work that a man can occasionally assist in alleviating, instead of seeing it as equally his responsibility (Kimmel, 200:128). This “helping out” or “pitching in” often occurs upon prompts or instructions, but this work of delegating household labour is also just that – work. Household labour often involves anticipating needs: replacing the toilet roll for the next person to use the bathroom, buying milk before the next person’s cup of coffee, noticing that there are items in the fridge that have gone off and need to be thrown away. If all the household work that a man does is delegated to him, then the work of noticing these needs still falls to the female partner, and this further enforces the idea that household work is a woman’s domain that is her responsibility to ensure it is done, whether it is by doing it herself or by delegating it, or asking for assistance. In dual-earner couples, women do vastly more housework than men do. For example, an Australian study revealed that male partners do on average 11 hours per week of housework whereas their female partners do about 23 (Bittman, 2003: 186). In an American study, men did 5 hours for every 20 hours that women did (Kimmel, 2000:128). That this division is due to gender is one hypothesis out of many. Other possible explanations for this division of work include the relative income of spouses, their relative time constraints, their particular gender ideologies, and the way that each partner constructs their own gender (Erickson 2005). The first is, perhaps, the most intuitive: the idea that division of household labour is determined by relative resources of the couple, rather than their gender per se. The idea is that whoever earns more money can “buy out” their portion of household labour (Erickson, 2005:339). However, this does not receive empirical support (Erickson, 2005:346; Bittman et al, 2003:209). There is some evidence that women do less housework as their pay increases, however, men’s contribution to household labour is not at all affected by their wives’ income up until the point of equality in income between the couple. This suggests that either more housework is outsourced, or if that is unaffordable, the housework simply does not get done. Hochschild found exactly this phenomenon in dual-earning couples (1979:205). Interestingly, at the point where the wife’s income begins to exceed her husband’s income, the husband does less housework (Bittman et al, 2003:207), and the wife does more.

12 Stellenbosch University https://scholar.sun.ac.za

Several sociologists theorise that this phenomenon can be explained by “power-balancing” (see Hochschild 2003a, Tichenor 2005 and Bittman et al. 2003). This is when the male partner compensates for his perceived loss of power due to earning less by doing less housework. Ensuring that the second shift remains his partner’s problem is thus a strategy to preserve dominance and relative status. This same strategy is followed by women: women who earn more than their husbands want to ensure that the second shift remains their own problem – they also engage in ‘balancing’ to preserve the structure of male domination in their relationship (Hochschild, 2003a:231). The gendered economy of gratitude confirms this: women who earn more than their husbands tend to be grateful, firstly, that they are able to contribute so much to the household coffers, and secondly, that their husbands allow them to do so (Hochschild, 2003a:212; 254). Men in those situations do not show gratitude for the greater income of their wives, nor for the extra work that their wives put into the second shift. Veronica Tichenor further found that higher-earning women are just as likely, or more likely, to defer to their husbands’ decision-making in order to preserve the gendered power differential (2005:192). The way in which men and women conceive of their earnings is also importantly different. Even when both partners earn more or less the same amount, they are likely to think of the male partner’s income as “essential” and the female partner’s as supplementary (Thebaud, 2010:334). Even in couples where the female partner earns more, the male partner’s work is considered more important, which is another reason that women do more housework in such situations (Thebaud, 2010:334, Tichenor 2005:193). This also contributes to women opting to reduce their work to part-time, or to stop working altogether, thus removing them from the dual- earner bracket, but this is a symptom of the same issues that characterise dual-earning couples. A further hypothesis is that of gender construction: if one conceives of oneself in masculine terms, one will do less household work as well as emotional labour, and if one conceives of oneself in feminine terms one will do more household work and emotional labour (Erickson, 2005:340). This is somewhat true for men – those who conceive of themselves in more feminine terms do perform somewhat more nurturing work (Erickson, 2005:345). However, this is not true of women – even those who conceive of themselves in more masculine terms are likely to do more caring work (345). Because they conceive of it in agentive and instrumental terms, it seems that

13 Stellenbosch University https://scholar.sun.ac.za housework is not merely an expression of identity or love for women, but rather, work – and women are held accountable to do it in a way that men are not (348). Men, on the other hand, do not conceive of emotion work in agentive terms but rather only as a part of interpersonal relationships – rather than seeing it as deliberate and considered work that they do, they see it as arising naturally and spontaneously from their feelings of affection or love. When something is seen neither as work nor even as deliberate, it is not seen as something that has to be done even if one does not feel like it, but rather as only done when one has the urge or it is convenient to do so. An alternative hypothesis is that of time constraints: whoever works fewer hours of paid work will work more hours of unpaid work. This, however, does not receive empirical support (Erickson, 2005:346). Yet another competing hypothesis is that of gender ideology – where a more traditional gender ideology will result in female partners doing comparatively more household work and male partners doing less than their respective egalitarian counterparts. However, there is only a significant correlation for men: a man holding a more traditional gender ideology will result in his wife doing more work and him doing less, but egalitarian and traditional women do the same amount of work (348). This all hints at the fact that the distribution of labour is not explicable in terms of the members of the couple having found a way of sharing that suits both of their lifestyles, their incomes, their ideologies, and the way that they would like to express their gender. Even when the effects of these other variables are accounted for, being female is still the primary predictor of doing more of the second shift (and all the kinds of work it entails). This may suggest to some that the difference is innate, but I will refute that notion using psychological studies later in this chapter. The sheer difference in hours spent on unpaid work between men and women will be argued in Chapter Three to be unjust due to exploitation.

1.2.2. EMOTION WORK

The division of what I have called emotion work is also gendered. Women do more emotion work than men do (Erickson, 2005:344). One of the characteristics I attributed to emotion work is that it must be received as such. However, there is a perception that women appreciate emotion work done for them, or others close to

14 Stellenbosch University https://scholar.sun.ac.za them more, or have more need of it. It may be true that women appreciate emotion work more, given that the economy of gratitude is premised on what people feel they can either legitimately or realistically expect from their partner. Nevertheless, it seems plausible to hypothesise that men and women have different expectations or experiences of emotion work - perhaps that women value it more, or that their experience of it is more positive. However, a recent study (Curran, McDaniel, Totenhagen and Pollitt, 2005) showed that for both partners, an increase in emotion work by their partner improved feelings of commitment to and satisfaction with the relationship. (Curran et al, 2015:169). There are some slight differences in the kinds of positive emotions engendered by being the recipient of emotion work by men and women, but this marked increase in relationship satisfaction is enough to show that emotion work in romantic relationships is important to both men and women. There is therefore already a hint that in order to maximise satisfaction with a relationship, a more equal division of emotion work is required. Positive marital outcomes are associated with equal emotion work (Curran et al. 2015:159), not with an overall level of emotion work. Further, men’s provision of emotion work ‘improved both their own and their wives’ marital outcomes’ (Curran et al, 2015:159). For this reason, there is no way to receive all the benefits of emotion work for a relationship or marriage without contributing to it equally. There is also a way in which men benefit from women’s emotion work for others indirectly. In my definition of emotion work above, I limited it to emotion work done for one’s partner. However, women also do more emotion work for people outside of their romantic relationships, and this benefits their male partners indirectly, so it is worth looking at here. The relevant social relationships include extended family, the community, and the friends of the couple. Examples of the effort involved include remembering birthdays and making a phone call, sending a birthday card, or sending a gift. It might also include cooking for someone (perhaps a neighbour) who is ill, or taking them flowers. It also entails making an effort to organise social events or family gatherings for the purpose of sustaining relationships. If the couple has children, this will also involve the children’s playdates and birthday parties, parents’ meetings and school events, and the birthdays of the children. These kinds of activities are, of course, not the entire stuff of relationships, but rather a deliberate demonstration of care, a way of enacting the value that a person has for the couple, the child, the family, or the individual.

15 Stellenbosch University https://scholar.sun.ac.za

It may be useful to argue that this can, in fact, be considered work. These interactions require effort, bookkeeping, planning, and attention. Especially in the case of the community and extended family, there may be other values in maintaining the relationships other than simply mutually enjoying spending time together, which means that often these interactions are perceived as effortful more often than friendships. However, even in the case of friendship, not every social interaction will be the product of natural enthusiasm. It seems to be more intuitive to women that friendship means putting in effort even when it is not convenient or even desired for its own sake – for example, attending birthday parties or weddings even at personal cost. Men, on the other hand, seem to be more loath to consider this work (and thus an obligation), because it seems to degrade the activity. This once again suggests that women are held accountable to this kind of work more than men are. If men do not see this as work, it is likely to be seen as non-obligatory, and to only take place when it is convenient and when the whim arises. I have not acquired data about the actual effort expended by male and female partners on people outside the partnership, but the evidence of this can be seen indirectly in the isolation of widowers. Widowers are left without a social network in a way that widows are not, which suggests that these widowers depended on their wives to maintain social connections. In a paper by Berardo (1970), aged widowers were ‘less likely to 1) be living with children, 2) to have a high degree of kin interaction, 3) to receive from or give to children various forms of assistance, 4) to have friends in or outside the community or to be satisfied with their opportunities to be with close friends’ (14). A further disturbing finding was that in the case of a surviving father rather than mother, the ‘frequency of interaction with extended family members was nearly as low as when both parents were deceased’ (Berardo, 1970:15). Further, the ‘loss of a spouse had a more adverse effect on the social participation of widowers than on that of widows’ (Berardo, 1970:16). As will be seen in the next section, social isolation is also very significant in terms of health. I will not refer to this kind of emotion work done for others outside the romantic relationships throughout this paper because, rather than being an issue of wrongfulness to the partner who does this work, I view it rather as the case of a loss of utility for the men who do not do this work. Being socially isolated without strong kinship relationships after the death of a spouse renders elderly widows vulnerable, which is a cost they bear, not their deceased wives. A further cost may be the

16 Stellenbosch University https://scholar.sun.ac.za meaningfulness of their relationships. This is an empirical question about how meaningful people perceive relationships where not much emotion work is being performed. However, emotion work does not only serve to engender feelings in someone else, but also serves to strengthen the relationship and cultivate intimacy. Without emotion work, it is at least possible that the relationship will be impoverished as a result. Once again, this is a cost borne chiefly by the men who lose out on the meaningfulness of relationships, if they do indeed experience them as less meaningful.

1.2.3. CARE

It is unsurprising that women are expected to do most of the caring work. In a study by Paoletti (2002), which focused on the care of older relatives with disability, it was found that caring is only expected from male relatives if there were no female relatives or if the female relative was for some reason unable to care (814). A study done on couples in Sweden in 2009 showed that women stayed home to care for sick children twice as often as their male partners, even though all parents in Sweden are entitled to paid leave to care for sick children (Eriksson and Nermo, 2009:344). Women thus take on the bulk of caring for the needs of other family members, even where structural impediments such as unequal or gendered paid leave have been removed. This is an indication of the extraordinary ideological hold of gendered patterns in heterosexual relations. However, for many marriages, the larger part of the health work that is done is maintenance and prophylactic rather than providing extra care due to illness, disability, or age. A wife managing her husband’s lifestyle for the sake of his health has become a trope in movies and literature. She might quietly warn him not to add extra salt, not to have an extra glass of wine, or a second helping of pork chops. If the burden of cooking daily falls to the wife, the burden of cooking according to a restricted diet will also fall to her – whether it’s weighing foods, doing research on what causes cholesterol, or what foods are and are not allowed for a diabetic. Some parts of the work that health requires are of course outsourced to health workers, but preventative measures such as diet and exercise, as well as actually accessing health care, is work that cannot be outsourced to someone outside of the couple. Women are

17 Stellenbosch University https://scholar.sun.ac.za far more likely than men to see a doctor both when they feel healthy (in other words, a check-up visit), and when they experience symptoms, and may also take on the work of ensuring their male partners go to the doctor in the case of illness or in the case of check-ups (Galdas et al., 2004:617). It may also fall to the female partner to ensure her partner takes his medication or follows lifestyle regiments. However, there are several reasons that this work should not fall to one partner only. All married people experience a health benefit in the form of social control, on which the management of chronic disease in part depends, since spouses are the most frequent source of social control for such management of disease for married people (August and Sorkin, 2010:1832). However, there is a gender discrepancy: whereas over 80% of men name their spouse as being an important source of influence and regulation, only 59% of women do (August and Sorkin, 2010:1832). That means that for more than 40% of women, their spouses are not an important source of social control for their health, which is more than double the men who do not receive this social control from their spouses. In this study, which looked at the management of type 2 diabetes, the married men in the sample received the most social control, and unmarried men received the least social control of all (unmarried women receive less social control than married women, but more than unmarried men) (August and Sorkin, 2010:1836). Further, though most people in the study experienced gratitude in response to social control, women experienced far more gratitude, which once again hints at who is held accountable for the provision of nurturing work, as well as what women feel they can expect from their spouses (August and Sorkin; 2010: 1837). This echoes the greater gratitude that women feel towards men for their contribution to household labour, whereas men feel comparatively little gratitude for the same. In general, ‘women possess more health-related knowledge, are more likely to monitor their own health status, and are less likely to engage in a number of risky health behaviors’ (August and Sorkin, 2010:1833). On the other hand, one qualitative study (O’Brien et al. 2005) showed that, for men, ‘there was a widespread reluctance to seek help (or to be seen to be seeking help) as such behaviour was seen as challenging to conventional notions of masculinity’ (514). This, however, places an additional care burden on the female partner.

18 Stellenbosch University https://scholar.sun.ac.za

A good example of this kind of burden is the account that this anonymous woman posted on a weblog of the responsibility she takes for her husband’s health after he is diagnosed with type 2 diabetes.3 She points out that she does this for selfish reasons, since she’d rather he didn’t die an early death, but here is her anecdote:

[…]on Monday, as my husband was getting ready to travel to an on-site job for a few days, I asked him if he wanted a protein shake with his breakfast, because I knew he had a karate class later that night, and wouldn’t really have much time to eat properly during the day […] There was some agonizing, and then yes, he wanted the shake. Awesome. Great. Then he said to me, as he was eating and I had sat down at the table to do some kind of crap household paperwork, that he wished he could avoid making decisions about what/when to eat. That he found it mentally tiring to be asked these questions and it would just be wonderful if I removed the element of choice entirely, and just gave him whatever it was he was supposed to eat. Because thinking about food all the time, when there are so many other things he needs to be thinking about is just too much sometimes. (user: skybluepink, page 17)

This account illustrates several useful things: firstly, the kind of burden that taking responsibility for someone’s health maintenance is. This woman has to consider her partner’s schedule for that day and determine in advance what his physical needs might be throughout the course of the day. The magnitude of the task and the mental energy it requires is laid bare by this example. Secondly, she has to take into account his preferences on an ongoing basis – she mentions earlier in this account that she adjusts what she feeds him based on whatever her partner is reading about diabetes at that moment. Thirdly, it illustrates how this man has an unquestioned assumption that his partner ought to take all responsibility for his health maintenance, where even his minor participation is too burdensome for him.

3 A weblog is an online forum where anyone can contribute a post. This one is particular was on MetaFilter, and people (mostly women) posted about their experiences of doing more emotion work.

19 Stellenbosch University https://scholar.sun.ac.za

The work of maintaining health falling to only one partner has some pragmatic and utilitarian costs. The first reason is that if the work is delegated to one’s partner, the skills required to manage illness, treat symptoms, or to take preventative measures are never acquired. This is a problem, firstly, because one’s partner is not always present, and simply cannot administer every insulin injection, or ensure every day that the daily statin is taken, or make sure that every meal eaten falls within a restricted meal plan. If the couple is separated, even if briefly, for example due to travelling separately, the male partner will lack the necessary skills to take care of his health when his female partner is absent. This is, firstly, inefficient: if such separations occur frequently, the health measures will simply be too infrequently enforced to be effective. Let us take high cholesterol as an example. If the meals at home are carefully monitored for cholesterol content, but the husband eats whatever he likes all through the day for his lunch and his snacks, then the restrictive diet will not be as effective as necessary in order to lower cholesterol. Further, if a wife is not present at every occasion where a statin is to be taken, it may result in statins rarely being taken, which also renders them ineffective. If the issue is a symptom that is not manifestly obvious, it is also simply impossible for a female partner to intuit these and insist her partner goes to the doctor. It may even restrict the caregiving partner’s freedom of movement if she is aware that the maintenance of health is dependent on her presence. If the work of health care (and possessing those skills that enable it) is the exclusive domain of the woman, it will also be the case that if her partner or another family member falls ill, it will be her responsibility to take care of her partner or children. This cannot be a reciprocal feature of the relationship – the male partner is likely to lack both the skills and the will (through lack of emotion work on the self) to administer caringly to his female partner should she fall ill (she will have to croak precise instructions from her sickbed). Since one of the goods of a relationship is sometimes being taken care of, at least when one is ill, this results in the female partner being denied a primary good of being partnered. Additionally, if a child falls ill and the father lacks the confidence in his skills (or, indeed, if the mother lacks confidence in his skills) in taking care of a child, it will, by default, be the responsibility of the mother to take care of the child, even if it means skipping work. This perpetuates the problematic fact of women being far more likely to make sacrifices in their careers in order to do family work. It may even result in a mother not quite trusting her husband to take care of the children when she is absent, if he

20 Stellenbosch University https://scholar.sun.ac.za cannot, firstly, notice when one of them is ill, and, secondly, know what to do about it. This is not to say that health should always remain solely each partner’s responsibility – as previously mentioned, one of the values of being partnered instead of single is that one does not always have to take care of oneself. For example, reciprocal tending to illness is desirable. Providing input when one becomes concerned about one’s partner’s health, or encouraging one’s partner to engage in healthier behaviours, is also appropriate. However, this should be reciprocal and mutual, and each partner must take final responsibility for his or her own health for the reasons listed above. Then, when one’s partner shows concern, it has the quality of a purer kind of emotion work rather than care work responding to an actual urgent need. These reasons are utilitarian and pragmatic, but the rest of the paper will also show that not only is it detrimental to men to distribute nurturing work unequally, it is also unfair for a female partner to have to singly take on these responsibilities per default.

1.3. THE BIOLOGICAL OBJECTION

I have looked at some potential explanations of the unequal distribution of work which are not merely gender-based, including relative time availability and relative income. However, it is necessary to refute one final claim that is made about the distribution of work which would make it an amoral discussion and therefore make the work of this paper impossible: that nurturing work is distributed the way it is because of the different innate abilities of men and women due to their biological sex. The discussion would become amoral because being able to do something is a necessary condition for being morally obligated to do it, and if men are not capable of doing nurturing work because of some feature of their biology, then they cannot be obligated to do it. Specifically, the idea must be considered that women are endowed with natural empathetic abilities, perhaps springing from their superior abilities in the use of language: women are supposedly better at discerning emotion, reading social cues and offering appropriate emotional support. This difference is supposedly ‘hard- wired’ into the female brain, whereas the male brain is simply less able to do such nurturing work (Fine, 2010:15).

21 Stellenbosch University https://scholar.sun.ac.za

In her excellent book, Delusions of Gender, Cordelia Fine refutes gender roles and characteristics attributed to biological sex. She examines studies which seem to indicate a gendered difference in empathising ability, notably the ones conducted by Simon Baron-Cohen. According to these studies, which included questionnaires where people rate their own empathetic abilities, women were far more likely to consider themselves empathetic than men. According to this self-reporting, at least half of women have an “E-type” brain (an empathetic brain – or, as re-dubbed by Baron-Cohen – a ‘female brain’). This is as opposed to only 17% of men who, according to this self-reporting, have a ‘female brain’ (Fine, 2010:16). However, as Fine points out, testing people for their empathising abilities through self-reporting is similar to testing mathematical ability by asking people if they are good at mathematics. In studies that actually test for empathetic abilities – for example, determining emotions from a picture of someone’s eyes, or attempting to read emotional subtext in an interaction, women do have a modest advantage. However, when a financial reward is offered (for example, $2 for every correct answer), men and women perform equally well in a test (Fine, 2010:21). Motivation can also be offered in the form of social value: if male participants are told that men with better empathising skills are more sexually successful and more likely to be seen leaving a bar with a woman they just met, men also perform much better on these empathising tests than without this information. It seems, then, that women consistently seem to be better at empathising because there is always a social motivation for doing so, whereas men need extra motivation. This is confirmed by the fact that men do seem entirely capable of doing emotional work at work. Psychotherapy is a classic example of such emotion work, and men are clearly able to do this when being paid to. Similarly, the emotional labour of maintaining social relationships which usually falls to the female partner in heterosexual relationships is effortlessly performed by men in the workplace in the guise of “networking”. When it offers financial reward, men implicitly understand that sending a gift or a Christmas card is a trivial gesture in itself, but is significant because of the work it represents in cementing and maintaining an important relationship. The extent of innate sex difference (in other words, hard-wired difference that is the result of biology) is, it seems, still an open question. There are many studies

22 Stellenbosch University https://scholar.sun.ac.za showing that hormones are determinant of behaviours, and studies that show the opposite, and studies that show that the ‘male brain’ and the ‘female brain’ are distinct and correspond to sex, and studies which show the opposite. However, I do not need to prove that there are no innate differences between men and women, and the above study is enough to show that, under the right circumstances, men are equally capable of nurturing work as women.

CONCLUSION

The aim of this chapter was to lay down the definitions of concepts I will be relying on in my arguments that follow, specifically explaining the constellation of different kinds of effort that make up nurturing work. I have argued that nurturing work has three elements which cannot be separated: housework, care, and emotion work. In the rest of my thesis I may refer to only one of these elements, since there are some points that are particular to, for instance, emotion work, but for the most part I will be referring to nurturing work. Secondly, I set out to show that the status quo of the distribution of labour between heterosexual partners is significantly unequal, and therefore the moral arguments that follow are pertinent to men and women in heterosexual relationships. The chapters which follow are theoretical, so all I need to argue is that if there is an unequal division of nurturing work between partners, it would be wrongful for the reasons I will present. Nevertheless, the empirical groundwork I have laid shows the practical relevance and indeed urgency of this work. Finally, I have attempted to eliminate objections to the idea that the unequal distribution of nurturing work is a case of gendered injustice. I have shown that gender is the greatest predictor of how much work one does in the (heterosexual) family, regardless of income, time, gender ideology, and the ways in which one perceives one’s own gender identity. This indicates that it is an issue of gendered injustice. I also excluded the biological objection because if valid, this would remove this work from the domain of moral theory: if men are biologically and physically incapable of taking on more nurturing work, there can be no moral obligation to do so. This is not the case – men are perfectly able to do nurturing work if the motivation is strong enough. As such, they can be morally obligated to do more work. The rest of this thesis will therefore treat the unequal distribution of nurturing work as a potential

23 Stellenbosch University https://scholar.sun.ac.za moral issue, specifically an instance of wrongfulness. The next chapter begins this project by determining whether it is a case of injustice, and if so, of what type, and thus whether it falls within the realm of political philosophy.

24 Stellenbosch University https://scholar.sun.ac.za

CHAPTER TWO: CONTRACT

INTRODUCTION

The previous chapter showed that the unequal distribution of labour cannot be justified by biology, nor by any other contingent factor such as free time or wage discrepancies. Given that the allocation of labour is virtually automatic according to sex, we should be suspicious that this may be wrong. Given that this affects most women and most men, we may also suspect that this is an issue of justice. As such, feminist liberals have used liberal theories of justice in order to frame this as a case of injustice. Further, they have attempted to use a liberal device to amend this: the contract. This chapter looks at two different strategies for employing the contract. The first attempts to use a contract between individuals as a means for testing for the presence of exploitation in a relationship. I will introduce this section by examining the individual contract and its role in the liberal tradition, as well as the marriage contract, using Susan Mendus. I will then look at how Jean Hampton appropriates this liberal tool for a feminist purpose. Hampton formulates what she calls a ‘contractualist test’ that draws on both Kantian and Hobbesian contractarianism and is meant to counteract tendencies that women have to be self-sacrificing. The second strategy is employed by Susan Moller Okin, and instead of using a hypothetical contract between individuals, she makes use of the Rawlsian social contract in order to argue for certain policy recommendations that she believes will rectify the issue of unequal distribution of labour; she is thus more concerned about institutional justice. I will argue that Okin’s is the better argument: her argument is theoretically sound and in keeping with liberalism, and more importantly, her policy recommendations are sensible and likely to be helpful for the distribution of household work and childcare. Nevertheless, her policy recommendations are unlikely to change any aspect of nurturing work apart from housework, and, crucially, it would not be possible to amend those recommendations to do so, since the theoretical justification for intervention in the form of regulation does not hold for other aspects of nurturing work. Hampton’s contractualist test may prove to have psychological

25 Stellenbosch University https://scholar.sun.ac.za value in the raising of feminist consciousness, but ultimately its formulation is such that it cannot test whether the distribution of nurturing work is exploitative. Feminist liberalism will thus be shown to have some use in establishing fairness in relationships, but nurturing work is beyond the field of its application.

2.1. MARRIAGE AND CONTRACT

2.1.1 LIBERALISM AND THE MARRIAGE CONTRACT

In her book, Feminism and Emotion (2000), Susan Mendus gives a useful overview of the liberal view of the marriage contract. Under a liberal framework, the way in which ‘private’ relationships (and here I am referring to relationships between citizens, usually business transactions) have been regulated is through the mechanism of the contract. The contract is justified theoretically by two aspects of liberalism: firstly, the notion of autonomy, and secondly, the notion of ownership of one’s body. Given that we are autonomous, when we freely enter into a contract, this is the source of the legitimacy of the agreement. The second justification is the notion that each person has ownership in her body, since no individual has any natural right over any individual except herself, and may contract on this basis, selling her labour or time. According to this justification, a contract must be mutually advantageous, with an equal distribution of costs and benefits, in order for the contract to be legitimate. Liberals have applied these notions to the marriage contract as well. Mendus explains that since liberals seek to legitimise power structures through the mechanism of contract, and since marriage is an instance of a power structure, it stands to reason that the contract can be used to legitimise marriage as well (2000:87). She however disagrees with this approach, since she find problems with both the notion of consenting to the marriage contract, as well as viewing the marriage contract as being justified by the fact that it is mutually advantageous. Firstly, she argues that the notion of autonomy used by liberals is unrealistic, which applies to all contracts, including marriage. Even if the contract is entered into freely, there are factors that could place the contracting parties in unequal positions relative to each other, such as differences in knowledge and power (91). Morally arbitrary factors will always interfere with someone’s decision to enter into a contract – ‘they may be bad judges of their own needs, bad judges of risk, or simply ignorance of the value of the goods being

26 Stellenbosch University https://scholar.sun.ac.za exchanged’ (91). For Mendus, the requirement of voluntariness is useful in determining whether or not a contract was freely entered into, which is important, but insufficient in determining whether or not the contract is legitimate. This is because voluntariness is not proof of autonomy due to the contingent factors such as differing levels of knowledge and power, and can therefore not by itself impart legitimacy to an agreement. Since autonomy cannot confer legitimacy, the terms of the contract must be (objectively) fair in order to make it legitimate (2000:92). Secondly, then, the marriage contract can be seen as being justified by the fact that it is mutually advantageous to the contracting parties. Given that the mere fact of voluntary consent is insufficient to legitimise the contract, the terms of the contract need to be fair; in other words, the costs and benefits need to be distributed fairly and equitably (93). This construes marriage as a kind of business relationship, where the parties strike a bargain. The first problem with this view is that marriage contracts simply do not work like this – the terms of a marriage contract are unalterable and externally defined, which was once hierarchical but is now technically non- hierarchical. The law, rather than the contracting parties, decides the terms of a marriage contract. For example, one can choose what kind of financial arrangement one prefers in a marriage by electing a marriage in community of property or outside of it, which can be amended with the addition of an antenuptial agreement. However, even these options are defined beforehand, and a couple is not permitted to strike a bargain unique to their agreement. Jean Hampton opts for a different route: she uses contractarianism not to alter the marriage contract, but to formulate a test for a good relationship. She draws on Hobbesian contractarianism as well as Kantian contractarianism to formulate a test for fairness in the terms of a relationship.

2.1.2 JEAN HAMPTON’S FEMINIST CONTRACTARIANISM

In her famous essay, “Feminist Contractarianism” (2007), Hampton draws on two kinds of contractarianism to formulate a test for the presence of exploitation in a relationship. She frames this as an antidote to what she sees as women’s tendency to be overly self-sacrificing in relationships at the unreasonable expense of their own interests. She says that her test can be used to evaluate any private, intimate

27 Stellenbosch University https://scholar.sun.ac.za relationship, including friendships, but she does place a special emphasis on heterosexual romantic relationships, and many of her examples are of wives and mothers. Her contractarian test is as follows:

Given the fact that we are in this relationship, could both of us reasonably accept the distribution of costs and benefits (that is, the costs and benefits that are not themselves side effects of any affective or duty-based tie between us) if it were the subject of an informed, unforced agreement in which we think of ourselves as motivated solely by self-interest? (Hampton, 2007:21)

This formulation will be unpacked once we have looked at the background that motivates this test. Hampton’s concern for women who are overly self-sacrificing is framed in terms of ethics of care. She uses Carol Gilligan’s interviews with children in which she found that the moral development of girls and boys differ – boys develop a justice-based ethics first, whereas girls develop a care-based ethics. Hampton uses two children that Gilligan interviewed as paradigmatic examples of each position, where Jake represents a justice-based ethics, and Amy represents a care-based ethic (2007:5). It is clear that the boy, Jake, errs on the side of acting perhaps too much in self-interest, where his idea of morality is ‘pursuing one’s own interest without damaging the interests of others’ (4), whereas for Amy, the girl, morality means putting others before oneself in apparently most scenarios: according to Amy, you should put your relationships before yourself ‘but to the extent that it is really going to hurt you or stop you from doing something that you really want, then I think maybe you should put yourself first’ (3, emphasis added). Hampton draws attention to Amy’s view that even in the case where one could really be hurt by putting others’ interests before one’s own, one should only maybe put oneself first. Hampton finds this a serious source of concern, especially if a woman like Amy ends up in a relationship with someone like Jake, where their respective ‘gendered’ moralities will make it likely that Amy will be exploited by Jake. Some feminists read Hampton’s concern about exploitability as being against ethics of care. For example, Ruth Sample (2002) suggests that if one agrees that doing care work is a virtue, women should not necessarily object to doing more of it (260). Further, if one frames the distribution of care work in terms of distributive justice, as

28 Stellenbosch University https://scholar.sun.ac.za

Hampton does, then it is being distributed as any other burden is, without being given special status as virtuous or celebrated – it may devalue care work in the eyes of proponents of ethics of care (Sample, 2002:261). Ruth Abbey, however, disagrees: she argues that Hampton is rather trying to present a more robust form of care ethics, where the consistent application of care means that the caregiver is obligated to care for herself as well as for others (2013:125-126). Caring for oneself is a requirement for healthy relationships, and so Hampton is prioritising relationships, including the relationship to oneself, which is characteristic of an ethics of care (126). The test that Hampton proposes is thus not privileging an ethics of justice over an ethics of care, but rather drawing on justice to strengthen and extend care. The theoretical justification for Hampton’s contractarian test draws on both what she calls ‘Hobbesian contractarianism’ and ‘Kantian contractarianism’. Hobbesian contractarianism, says Hampton, is characterised by total self-interest. That is, the contract imposes no moral obligation, but it is in the contracting parties’ self-interest to enter into the contract (11). The contracting parties will then insist on not being exploited by the contract, since this would be against their self-interest. For Hampton, the advantage of this position is that it involves little metaphysics – it is a simplified ‘morality’ where morality is reduced to mutual advantage. In contrast, Kantian contractarianism, says Hampton, is a mechanism to ascertain moral principles, and as such does impose moral obligations (14). In general, she prefers the Kantian position, and this is because Hobbesian contractarianism leaves no room for the intrinsic worth of persons. One only cooperates with others insofar as it is beneficial to oneself, and as such, people can only have instrumental value (11). This she considers to be a failure in capturing the nature in morality. Kantian contractarianism, on the other hand is premised on the intrinsic moral worth of persons, and for this reason is a much better account of morality. Hampton nevertheless will retain one element of Hobbesian contractarianism: acting in self-interest and making sure one gets what one deserves in interactions with others. Hampton chiefly uses Rawls as an example of Kantian contractarianism, and in doing so addresses the most common feminist critique of Rawls. For Rawls, the social contract is designed from the Original Position, which is behind the ‘veil of

29 Stellenbosch University https://scholar.sun.ac.za ignorance’4. This entails that everyone is stripped of their socially defined identities in order to find a ‘transcultural truth’, but feminists argue that our being is so rooted in social relationships that ‘easy intuitions remaining after people are supposedly stripped to their bare essentials will still be permeated with the assumptions of a sexist society, producing (not surprisingly) “patriarchal outcomes”’ (14-15). This occurs in two ways: firstly, these biases are likely to be present in the very assumptions and constraints that form the background to the formulation of the contract. For example, the fact that people in the original position are expected to make decisions based on self-interest and while disregarding emotion-based attachments to other people could be an instance of a masculine bias that the veil of ignorance would not counteract (16- 17). Secondly, if it is the case that masculine biases are pervading the original position process, then it is possible that they influence or determine the outcome of the original position procedure. If this is the case, and the maximin5 principle cannot be justified by the original position argument, then all we are left with are ‘vague intuitive appeals’ to the maximin principle (17). Hampton is not extremely sympathetic to this feminist critique - she argues that an ‘Archimedean point’ which is not entirely explicable by context is necessary for feminists as well: ‘they not only want their values to predominate, they want them do so because they are the right values’ (15-16). Hampton thus finds existing contractarian theories of both the Hobbesian and the Kantian traditions unsatisfactory from a feminist perspective: she dislikes the masculine bias in the Kantian traditions (including the Rawlsian version), and dislikes the instrumentalisation of people necessitated by the Hobbesian approach. Nevertheless, she finds the idea of contract an appealing tool which, if amended, could become a feminist device. Hampton believes that the contract can be used to

4 Rawls proposes the Original Position as a means to determine the principles of justice according to which we should organise society. He suggests that this takes place behind a veil of ignorance, where everyone is stripped of their defining characteristics, and could theoretically occupy any position in society. As such, people will decide in self-interest, but which is not based on their identity - in other words, it is the self-interest of the anonymous citizen rather than an existing citizen with partisan interests. 5 The maximin position is one of the ways in which people will decide in the original position: because one might reenter society as anyone, including the person who is worst off in that society, those in the original position would, according to the Rawlsian wager, choose to maximise the minimum that anyone in society gets, so that the worst off in this societal structure would be better off than in any other societal structure.

30 Stellenbosch University https://scholar.sun.ac.za make ‘a moral evaluation of any relationship’ including those in the private sphere, such as that of family and friendships, as well as romantic relationships. The moral evaluation is specifically to ascertain if the relationship is just. She does not define what she means by justice, but she makes it clear that at the very least, it means the absence of exploitation. Once again, she does not make explicit what she means by ‘exploitation’, so I will assume she is using the standard definition: ‘A wrongfully exploits B when A takes unfair advantage of B’ (Wertheimer, 1996:12). Instead of seeing love as voiding the need for justice, as many political philosophers have, including canonical political philosophers such as David Hume and Jean-Jacques Rousseau, Hampton sees the presence of affection and duty instead as precisely the point of vulnerability which can be exploited. As she puts it, ‘one’s propensity to give gifts out of love or duty should not become the lever that another party who is capable of reciprocating relies upon to get one to maintain a relationship to one’s cost’ (2007:21). Hampton’s paradigm example of this occurring is a mother whose love for her family is used by her family to ensure that she does extra work for them without reciprocation or compensation. The fact that she works without receiving anything in return means that she bears all the costs, and those whom she loves and does work for receive all the benefits, which makes it a losing situation for her. Now let’s return to Hampton’s formulation of her contractarian test. The test demands that the distribution of costs and benefits be evaluated for fairness and equitability, but she excludes costs and benefits ‘that are not themselves side effects of any affective or duty-based tie’ (2007:21). Presumably, a duty-based tie might be something like a parent-child relationship, where the distribution of costs and benefits are typically the result of the duty one has to care for one’s child. Further, though she refers to an ‘affective or duty-based tie’, it seems that she actually means that the affective side effects of a relationship should not be taken into account. I say this because the examples she provides are as follows: ‘the pain a parent feels when her teenage child gets into trouble, the happiness felt by someone because of the accomplishments of her friend, the suffering of a woman because of the illness of a parent’ (21). These all seem to be examples of positive and negative emotions that are felt due to the intimacy of the relationship. Therefore, I will assume that it is positive and negative emotions that are being left out of the calculation of the distribution of costs and benefits.

31 Stellenbosch University https://scholar.sun.ac.za

Hampton justifies her exclusion of affect and duty firstly, because she believes that these can blind one to what is in one’s self-interest. She believes that it is precisely one’s love and affection for others that it taken advantage of – that is used as the ‘lever that another party […] relies upon to get one to maintain the relationship at one’s cost’ (2007:21). Her key example is that of the housewife whose family uses her love and affection for them to get her to do more work for them than they do for her. Secondly, she states that affection and duty ‘cannot be distributed and are outside the province of justice’ (21). She suggests the positive and negative emotions felt due to intimacy cannot be distributed to someone else (for example, one cannot and presumably would not want to redistribute the concern a mother feels for her child). Further, the test she is developing does not test for equality, but instead whether someone could reasonably agree to the distribution of costs and benefits. As Linda Radzik explains, for Hampton, this need not involve optimising one’s situation so that one can receive the maximum benefits given the situation, but rather, the distribution should merely be such that a reasonable person could agree to them (2005:49). Hampton touches on the role that consent plays in her vision of the contract, but argues that consent could only truly occur if the terms of the contract are fair (26). As we have seen, consent and fairness of terms (in other words, whether or not the contract is mutually advantageous) are two separate criteria for giving a contract legitimacy, which have different theoretical bases in liberalism. However, Hampton frames her contract is an ideal one rather than a morally neutral device, which means it is ‘fed by normative ideas that one is ultimately relying on when using the test to make moral evaluations’ (26). This means that the idea of intrinsic worth built into the very idea of the contract, and if it is not present, it cannot be this kind of ideal contract. For this reason it would be impossible, in principle, to freely or autonomously consent to a contract unfair to oneself because it disrespects one’s intrinsic worth. According to a Kantian perspective, says Hampton, a contract that is not at least minimally mutually advantageous necessarily entails an implicit disrespect for the losing party. Hampton further emphasises that the Kantian respect for persons also entails equal respect for oneself (23). One cannot therefore autonomously consent to a contract that degrades one’s moral worth.

2.1.3. CRITIQUES OF A CONTRACTARIAN TEST FOR INTIMATE RELATIONSHIPS

32 Stellenbosch University https://scholar.sun.ac.za

Abbey argues that the device of the contract in Hampton’s theory is purely metaphorical and does no normative or theoretical work (2011:128) She argues that Hampton’s theory is simply Kantian feminism, where the introduction of Hobbes as well as the device of the contract are superfluous. Firstly, Abbey suggests that Hampton relies so little on the contract that her position might more aptly be called ‘Kantian feminism’ rather than ‘feminist contractarianism’. Hampton states that her ideal form of the contract means that one can only consent to a contract that respects one’s intrinsic worth. Of course, not all contracts function this way, such as the Hobbesian kind of contract that instrumentalises others, or an employment contract that exploits the employee. Hampton uses the meaning of contract univocally, in such a way that it is synonymous with a Kantian respect for persons. If we recall Mendus’ distinction between autonomy and fair terms as different ways of legitimising contract, it is clear that Hampton only acknowledges the latter, since she believes ‘true consent’ is predicated on fair terms. Abbey also argues that the inclusion of Hobbes into the theoretical justification of Hampton’s formulation of the contractual test is not only superfluous but possibly also harmful to her argument. It is superfluous because, Abbey argues, one does not require the explicit inclusion of self-interest in order to achieve the kind of justice Hampton is after (i.e., distributive justice of non-affective costs/benefits). The Kantian position that everyone has intrinsic moral worth naturally includes the woman in question herself, and so already implies an obligation to consider oneself as equally important as people with whom one is in relationship (2011:129). Hampton includes Hobbes so that the woman in question will insist on not being exploited, but exploitation (if it can be considered treating people as means to an end instead of ends in themselves) is already prohibited by the demand to respect others’ intrinsic worth (129). In fact, argues Abbey, the inclusion of Hobbes risks misrepresenting Hampton’s position and making it unappealing to feminists: since she explicitly includes self-interest as the test for exploitation, it may seem that she is instrumentalising relationships in the same way Hobbes does (130). Hampton does not actually do this, given her emphasis on intrinsic rather than instrumental value, and moreover she states that there can only be real love in a relationship if there is also respect for the intrinsic value of one’s friend or partner. She also acknowledges that people’s motivations for entering into a relationship with someone is neither

33 Stellenbosch University https://scholar.sun.ac.za instrumental, nor in order to pursue justice, but have to do with different factors altogether (such as love). If Abbey is right, Hampton’s theory could be streamlined by leaving out Hobbes altogether. However, Hampton is not simply making an abstract philosophical argument: she has identified a tendency in women which is socialised rather than natural, and it seems to me that she is trying to provide a tool to counteract that psychological tendency. Let us examine the alternative (stripped down) formulation to Hampton’s contractarian test that Abbey provides, given the above arguments: ‘could we accept this [relationship] if we think of ourselves as beings with intrinsic worth, whose worth warrants respect from all humans capable of extending it?’ (130). This formulation excludes an evaluation of the distribution of costs and benefits, and therefore whether self-interest would dictate that one ought to stay in the relationship because the (non-affective) benefits outweigh the costs of being in the relationship – in other words, because one gets at least as much as one gives. Rather, one asks whether one’s intrinsic worth is respected by the relationship, and the only way to determine this is to ascertain if the relationship is exploitative. Abbey points out that this contractarian test is an act of feminist consciousness-raising that allows a woman to get the required distance from relationships so that she can test them for exploitation (2011:127). If she is right, and I think she is, then the test serves a primarily psychological function, rather than positing a moral position. Therefore, though Abbey’s formulation may well more simply and accurately represent Hampton’s Kantian feminism, it would not serve the psychological purpose equally well. If a woman were to ask herself Hampton’s version, she may well acknowledge that the distributive arrangement of burdens in her relationship is contra her self-interest. According to Abbey’s formulation, however, this means that she would also have to admit that her relationship is unacceptable because her intrinsic worth is not respected or recognised. The latter seems to be a far more drastic conclusion than the former – it seems to imply vicious sexism rather than a lack of distributive justice, and may thus not ring true to the woman who applies the test to her relationship. The deliberate inclusion of self-interest in the test is a way to ensure that one’s intrinsic worth is respected by repressing the tendency to be self- sacrificing in a relationship, since women are socialised to put the well-being of others and the relationship above their own needs.

34 Stellenbosch University https://scholar.sun.ac.za

Although the point has been argued above that love does not void the demand for justice, the presence of love nevertheless complicates the question, ‘does my partner value my intrinsic worth?’ Because it is an intimate relationship, it is highly unlikely to be merely instrumental – if a man viewed his partner entirely instrumentally, he would view her as replaceable by any woman who could perform the work equally well. It is improbable that many (if any) relationships are actually such a brute bargain. It is more likely that a man loves his wife for her particular characteristics and personality, which means that he recognises her intrinsic worth. This may not mean that he respects her in the ways in which Kantian morality would demand given her equal intrinsic moral worth, which, Hampton suggests, could never allow exploitation. Nevertheless, since the answer to Abbey’s formulation sounds much more vicious, which would be that the problem with the relationship is an absence of respect for intrinsic worth, it is less likely to be helpful to a woman employing it. This is not a philosophical point, but rather a psychological one. Hampton’s moral position can be summarised as follows: given the equal worth of persons, one must recognise one’s own worth and realise that this imposes a moral obligation on others to treat you with respect, and on you to insist on such treatment. Hampton emphasises the woman’s perspective and obligations in this scenario: women have a moral Kantian duty to not allow themselves to be exploited. However, it is possible to extrapolate from her Kantian perspective that exploitation is forbidden, and in order to respect the intrinsic and equal moral worth of persons, men may also not acquiesce to a distributive arrangement that is exploitative of their wives, and in fact it seems wrong-headed to focus on individual women ‘allowing’ their partners to disrespect their intrinsic worth. What is problematic about this is that it makes the party more likely to be exploited responsible for rectifying the situation. In fact, Ruth Sample makes a convincing argument for why focusing on an individual woman’s assessment of whether or not her relationship serves her self-interest is nonsensical without taking societal conditions into account. In her paper, Why Feminist Contractarianism? (2002), Sample provides some critiques of Hampton’s position, the most compelling of which is her argument that the exclusion of affective and duty-based ties is firstly, unjustified, and secondly, would imply the destruction of the family, as well as her argument that Hampton’s paper unfairly makes women seem irrational for being in exploitative relationships.

35 Stellenbosch University https://scholar.sun.ac.za

Sample offers an alternative account for when and why these kinds of relationships could be considered exploitative, according to her own definition of exploitation: when someone gains an advantage by disrespecting someone else’s intrinsic worth (2003: 56) Sample argues that the content of Hampton’s contract excludes too much: she suggests that including affective ties and duty is essential in calculating the utility of an existing arrangement. She suggests that the affect involved in nurturing work and household work could make the work not only tolerable but even joyous (269). This may well be the case for emotion work – as I have mentioned previously, emotional labour usually springs out of love (which nevertheless does not exclude it from the province of work). This is also plausible for childcare, naturally, but less plausible for housework, particularly cleaning. Nevertheless, given that, realistically, a woman’s options are either to have a family and have the benefits of that come from affective ties associated with the family under conditions of unequal work, or to not have a family, if a woman wants a family, she is rational in choosing to have a family under those conditions. It may be the case that a man making the same decision to opt into family life gets more utility from that decision, but given that a woman does not simply have the option of acquiring the same utility as a man, her decision maximises utility for herself, given her options, and therefore is a rational choice. It is important to note that Sample is not arguing that this makes the conditions of the relationship fair, but rather a woman’s decision to opt into that relationship, or to stay in that relationship, rational (269). So why does Hampton exclude affect and duty-based ties? She argues that they ‘cannot be distributed and are beyond the province of justice’ (Hampton, 2007:21). Sample provides multiple counterarguments to this position. Firstly, is it the case that affective benefits cannot be distributed? Sample argues that since these benefits are often associated with roles, and roles can be distributed, it seems that affective benefits and benefits from duty-based ties can be distributed (2002:270). Since emotion work deliberately aims to evoke positive emotions in one’s interlocutor, and both partners can do emotion work, it seems that at least some positive emotions one feels are strongly associated with emotion work done by the partners. The exclusion of affect and duty-based ties should hint at the fact that Hampton’s contract would be useless in distributing nurturing work. It seems that, at

36 Stellenbosch University https://scholar.sun.ac.za least much of the time, women would take responsibility for the health of their husbands out of love or a sense of duty, and the same would apply to doing the emotion work of feeding egos and tending wounds. Sample suggests the kinds of sacrifices made for one’s family can only be explained in light of one’s love for them, and in the case of duty-based ties such as a parent-child relationship, one’s duty to them. This is patently the case with childcare: if one were to disregard both one’s love for one’s children and one’s duty to one’s children, and to act only in self-interest, then no sacrifice as a parent could be intelligible. Further, if, for example, one makes sacrifices in one’s own career in order to support a move in one’s partner’s career, this would certainly not be explicable in terms of self-interest, but rather only in light of one’s affective ties or duty. This fact leads Sample to argue that no family would survive the test of the contract, since much behaviour in the family can only be explained by affective ties and duty rather than by self-interest; this might even be what most pertinently distinguishes family life from public life. For example, if a woman is offered a promotion that involves moving to a different country, and her husband leaves his current job for a lower-paying job in the new country in order to support his wife’s career, he would suffer the costs of giving up his current job, as well as being separated from his friends and family and everything he is familiar with. With costs this great, it is hard to see what benefits would need to accrue to this man in order for him to make this decision out of pure self-interest. Of course, presumably he is not making this decision out of self-interest, but out of love and support of his wife. His behaviour cannot be explained by self- interest, but is neither obviously irrational nor indicative of an exploitative relationship. Thus, according to Sample, if we apply the test strictly, it would ‘demand the virtual abolition of the family’, since the kind of sacrifices we make are overall not in our self-interest and cannot be explained in those terms (2002:272). Furthermore, it is not clear that this relationship is unfair to the male partner because he has elected to make this sacrifice for his wife. It thus seems that, firstly, the kinds of decisions we make out of love cannot be tested using purely self-interest, and secondly, that one partner sacrificing more than the other is what makes the relationship unjust. In other words, there must be a reason other than an inequality of relative costs and benefits accruing to the respective partners that make a relationship unjust..

37 Stellenbosch University https://scholar.sun.ac.za

If we change the hypothetical case study to swap the gender roles, our intuitions may be slightly different: we may be more suspect of the justice of the situation if the female partner was the one who gave up her job to move with her male partner to a different country. This is because we suspect that this decision was less free because a woman typically faces greater restrictions in terms of her career purely on the basis of being a woman. Sample argues that, when women in relationships constantly sacrifice their own self-interest for the good of others, it is exploitative, but not because their preferences are irrational. Rather, it is exploitative because it takes place against an extra-familial, social, public background of injustice. Her preferences are not the natural outcome of her personality, her biology, or her reasoning, but the outcome of a socially constructed sexual inequality. Sample argues that the fact that women, on average, bear more of the costs and fewer benefits than the men with whom they are in relationships, is unjust because it is the product of oppression. Those ‘preferences’ into which women are socialised occur the way they do because they serve the patriarchy and detriment women. The options available to women are fewer than men (for example, a man can have all his childcare and chores done while still having a career, while usually women have to make sacrifices on both to have either), and this is why it is exploitation. Sample argues ‘gendered expectation and mutual awareness’ is what causes oppressive relationships:

Because men expect that women will be more likely to act out of affection and loyalty, and because men are correct in that expectation, and because women expect that men are more likely to act out of self- interest, and because women are correct in that expectation, this provides men with greater bargaining power within the family. (266-267)

Sample is thus arguing that placing the contractarian test in the hands of an individual serves no purpose, because the way they ‘choose’ to design the distributive arrangement of their relationship is not really up to them. Rather, the way they have been socialised into their genders as a result of an unequal society, and the fact that this determines that there is often exploitation of the female partner, is the reason for injustice. Sample suggests that contractarianism may still be a useful tool, but only if

38 Stellenbosch University https://scholar.sun.ac.za we use it to create a more just society. The next section explores exactly this: how can the social contract be used to make the family more just?

2.2. THE SOCIAL CONTRACT AND THE FAMILY

2.2.1. APPLICABILITY OF JUSTICE TO THE FAMILY

Susan Moller Okin is a well-known feminist liberal who argues in her book, Justice, Gender and the Family (1989) that the family cannot be excluded from the purview of justice. Traditionally, political philosophy has seen the family as lying beyond justice. Jean-Jacques Rousseau, for example, argues that excluding women from political participation would not result in their interests not being taken into account – since men love their wives, they would always act in their best interest (Okin, 1989:26). Hume argues that justice is an inappropriate virtue for the family, since superior virtues rule the family (27), such as ‘spontaneous affection and generosity’ (28). However, Okin points out that the presence of ‘superior virtues’ does not exclude the necessity for justice: she agrees with Rawls, who names justice the primary virtue because it needs to be present in order for other virtues to be present. Further, this idealised vision of the family is inaccurate, or at least not the full picture: the prevalence of domestic violence indicates that spontaneous affection is not necessarily the rule that dictates all families. Okin also points out that what supererogation occurs in a family usually happens at the expense of the female members, where they are more likely to be self-sacrificing and not to claim what is ‘due’ to them (31), which is exactly what Jean Hampton tries to address with her contractarian test. In the second place, she points out that the idea that the law has in fact left the family alone, and that feminism’s demand for intervention into the family is novel, is a convenient myth. Historically, the law has deliberately maintained the patriarchal structure of a marriage by defining it in hierarchical terms, and depriving women of their income and property once they are married. In the present day, the law dictates who may and who may not get married, what constitutes a family, when people may get divorced, and so on. As Okin points out, ‘the issue is not whether, but how the state intervenes’ (131), to shape, create and limit the ‘private sphere’.

39 Stellenbosch University https://scholar.sun.ac.za

Thirdly, Okin argues that the family is the site where we are first socialised into our genders. This is a high-stakes activity, given the lower status that women occupy. Okin relies on Nancy Chodorow’s paper, “Family Structure and Feminine Personality” (1974) to argue that the traditional gender division of labour in the home creates significant gender differences. Chodorow posits that the fact that children’s primary parent is most often their mother or a female surrogate affects the gender development for boys and girls. Girls are encouraged to identify with their female parent, and because she is present during their development, they develop personality characteristics that make them empathetic, good at nurturing, and psychologically connected with others (Okin, 1989:131-132). This makes them both suited to and interested in nurturing, and therefore also more likely to choose nurturing occupations (or to invest most of their time in their families) in adulthood. Boys, on the other hand, are encouraged to identify with their male parent, who is away for the greater part of the day engaged in activities the children know nothing about. As a result, boys prioritise autonomy above connection, and aspire to ‘public status’ rather than to be good nurturers (132); in fact, the domestic, maternal sphere is turned into that which they must transcend in order to become a man. Finally, the consequences of the division of labour in the home are not limited to the home, but rather permeate every level of a woman’s life (132). Their resulting absence (or at least comparatively very small presence) in positions of authority or high-status occupations6 means that women are still not associated with authority and competence, which also means that it is more difficult for individuals to break the status quo and pursue such positions as women. In fact, Okin argues that the vulnerabilities that are the consequence of the division of labour are grounds for regulation of the family – in other words, one of the reasons why justice applies to the family and why government regulation is necessary. The ways in which women are made vulnerable by marriage will now be explored, and then I will explain Okin’s

6 In South Africa in August 2017, ‘women comprised 32% of Supreme Court of Appeal judges, 31% of advocates, 30% of ambassadors and 24% of heads of state-owned enterprises. If we take a brief look at the Top 40 JSE listed companies, only one company had a female CEO.’ http://www.statssa.gov.za/?p=10325

40 Stellenbosch University https://scholar.sun.ac.za argument that this constitutes grounds for (different) intervention into the family, and finally Okin’s policy recommendations for such intervention.

2.3.2 VULNERABILITY BY MARRIAGE

As previously mentioned, political philosophy has traditionally considered the family as being beyond the reach of justice. According to Okin, liberalism has only obliquely addressed the family: firstly, the family one is born into can act as an impediment to equality of opportunity, but this is only considered on an interfamily level – the inequality of opportunity between different families is of concern to the liberal paradigm (134). Secondly, it was considered a domain where justice was inappropriate, and moreover unnecessary - since the family is ruled by love, it has no need of justice. Finally, it has been considered as a school of justice, where future citizens are cultivated with a sense of justice (134). However, Okin believes that the family is currently a site of injustice, where the essential goods are unevenly distributed, including power, prestige, self-esteem, opportunities for self- development, and both physical and economic security (136). For Okin, the division of labour within the household is clearly a question of justice because, 1) labour in heterosexual couples is nearly automatically allocated on the basis of an irrelevant innate characteristic, which is the kind of discrimination theorists of justice are typically interested in; 2) this automatic allocation persists throughout the years in which the couple have children, and the distribution remains unequal beyond that point – the problem is thus more or less permanent, rather than temporary; and 3) its impact on partnered women is so great that ‘it affects every sphere of [women’s] lives, from the dynamics of those marital relationships to their opportunities in the many spheres of life outside the household’ (Okin, 149). The impact specifically results in women being made vulnerable, and, as we will see, Okin takes this vulnerability to impose a special moral obligation to regulate marriage. She identifies three kinds of vulnerability associated with marriage: vulnerability in anticipation of marriage, vulnerability during marriage, and vulnerability due to separation or divorce. The first kind of vulnerability is due to the kinds of choices girls and women make based on their expectations that, firstly, they will get married and have children, and secondly, that they will be the primary parent (142). Because of these expectations, girls are less likely than boys to pursue

41 Stellenbosch University https://scholar.sun.ac.za prestigious careers and less likely to prioritise financial independence (143). On the other hand, they are more likely to opt for careers that lend themselves to flexibility, or jobs that do not hold promise for career advancement (see Wiswall 2016). Boys conversely expect to be the primary breadwinner (and not be otherwise impeded by family responsibilities), and so make decisions for prestigious careers that entail advancement. Girls feel that they have to choose between having children and having careers, whereas no such dichotomy exists in the minds of boys. These facts that Okin brings up may seem to have aged more than her other data, but recent data from Sweden, which ranks fourth in the world in gender equality (World Economic Forum, 2016:328), show that occupations involving care work and little chance of career advancement are far more likely to be occupied by women, such as being nursing assistant, personal care assistants, cleaners, and secretaries. There are parallel jobs mostly occupied by men such as truck driving, but there are fewer such jobs than those occupied by women (Statistics Sweden, 2016:57). Further, in a recent TED talk, the CEO of Facebook, Sheryl Sandberg, advises women not to start making career sacrifices until they are actually pregnant (2010). In her experience, women are likely to begin turning down career opportunities in anticipation of children long in advance of having children (see Strober, Miling and Chan 2001:132). When examining the vulnerability of women in marriages, Okin distinguishes between the case of housewives and of wage-working wives. Firstly, housewives do work fewer total hours than their husbands, but all of their work is unpaid (though they do receive a financial benefit in the form of maintenance). Okin lists the disadvantages of being a housewife as follows: much of the work is boring and unpleasant, her work is unscheduled (which means that sometimes it never stops), she cannot change jobs as easily, her lack of income means that, at best, she has less bargaining power than her husband, and, at worst, is dependent on the charity of her husband. She acquires no status from her work, but her status rather depends on that of her husband. Most dramatically, however, is that being a housewife makes it much more difficult to exit a relationship. As such, this makes them more vulnerable to abuse, and to tolerating unacceptable relationships in general (149-152). It is somewhat easier for a working woman to leave a marriage than housewives, but her lot during the marriage is arguably worse than that of the housewife. Whereas housewives work fewer hours than their husbands (paid and unpaid), working women work far more (see chapter 1), where the largest disparity is in the unpaid work that

42 Stellenbosch University https://scholar.sun.ac.za they do. As such, the hours that she can spend on her paid work are far less than her husband can, which hinders her career advancement as well as her leisure opportunities. The vulnerability of women after separation or divorce is perhaps the most significant, particularly that of housewives. The career decisions made by both housewives and working wives during their marriage were likely rational given their marriage and relatively limited career options, but tend to be highly detrimental if they became single. Since being a housewife is not considered to lead to the acquisition of marketable skills, housewives are far less employable than almost any other group. Women who are wage earners but have made sacrifices in their careers for their families, such as working part-time, or giving up opportunities at work in order to be more available for their families, are also at a disadvantage compared to their husbands. As such, even if the material assets of the marriage are divided equally, the most important asset is the male partner’s earning power. This, in addition to the gender pay gap, where women still earn 78 cents to the dollar, means that women are almost guaranteed to be financially worse off than their husbands. In addition to this, women are far more likely to be custodial parents (82,6% of custodial parents are women) (Grall, 2007:2), which means that their day- to-day expenses are greater due to their larger households. Further, the day-to-day care of children continues to disadvantage them in their careers in a way that noncustodial parents are not disadvantaged (Okin, 1989: 162). This is all borne out by a statistic from 1986 that Okin mentions: in the first year after separation, the standard of living of the male partner increases by 42%, and the that of the female partner drops by 73% (161). A statistic from a decade later, in 1996, indicates that the boost experienced by men is probably closer to 10%, and the drop experienced by women is 27% (Peterson 1996), which is nevertheless significant. Further, a recent study from the United States indicates that custodial parents are twice as likely to be below the poverty line as anyone else (Grall, 2007:4). Thus, especially separated women are far worse off than women who have never been married, and certainly than their ex- partners. Okin depends on Goodin’s argument in his book, Protecting the Vulnerable (1985), that we have an obligation to protect the vulnerable. He claims that the special obligations we have to particular people are typically conceived of in terms of voluntarism – we have a duty to them because we elected to be in relationship with

43 Stellenbosch University https://scholar.sun.ac.za them. He counters this by saying that our special obligations come from the vulnerability of those close to us rather than from the voluntariness of our relationship with them (Goodin, 1985:42). This argument applies both the interpersonal relationships and to the obligations of the state. Goodin argues that in interpersonal relationships, our intimates are vulnerable to us by virtue of their closeness to us, and for the same reason, we are in a particular position to protect them (1985:115). This gives us a duty to give them special consideration. He extrapolates this the obligations of the state: the moral community as a whole has a duty to protect the vulnerable, and as we share this duty collectively, we ought to ‘see to it that a cooperative scheme is set up and operates effectively to protect all the members of the community’ (153). This can be outsourced to the state, which, because of its relatively massive resources, has the greatest obligation to protect the vulnerable (153). Along with this, where vulnerability is socially created rather than a natural feature (like that of the infant), we must reduce vulnerability as far as possible. Given that we have an obligation to mitigate vulnerability where possible, Okin argues that different regulation of marriage is justified. Gender-structured marriage is a necessarily evil. We cannot prevent people from choosing it, but we must mitigate the vulnerability that is the consequence of that choice. To do that, marriage must be subject to the following regulations: firstly, both spouses should be equally entitled to the total income of the household for the duration of the marriage (Okin, 180). Okin suggests that half a person’s paycheck should be made out to their spouse. Okin suggests that for most couples this will simply formalise what is already the arrangement between a couple. Yet she hopes that this will rectify a power imbalance in a relationship, because it symbolically asserts that unpaid work is just as important as paid work, rather than attempting to compensate the home-making partner for services rendered (181). Secondly, divorce laws need to be changed so that women are protected from the vulnerability associated with divorce (183). The prescribed alimony and child support should be enough to ensure that both households enjoy the same standard of living, and should endure at least as long as the marriage did rather than only being considered temporary and rehabilitative (with the underlying assumption that the custodial parent receives an independent income) and thereafter the child support should ensure that the child has the same standard of living as the noncustodial parent.

44 Stellenbosch University https://scholar.sun.ac.za

Okin also suggests that the workplace and schools would need to be reformed. Firstly, the state must provide high-quality subsidised daycare, and large employers should provide on-site daycare (176-177). Secondly, workplaces will need to assume that many employees have responsibilities to care for children or elderly relatives and make provision for this. There must be parental leave for both parents, and both parents must be able to work part-time or flexible hours until their child is of school age. Thirdly, schools need to provide high quality after-school care so that they do not operate on the assumption that there is always a parent at home in the afternoons (177). Okin argues that these policy changes will help to equalise power relationships between men and women, and ensure that the exit costs are more or less the same for partners in relationships. Okin also believes that these policy changes will ensure that women and men are equally represented on every level of work and politics, and thus the benefits of a more equal private sphere will significantly benefit public gender equality (184). Further, children would benefit as well: since custodial parents are more likely to be below the poverty line, so are their children. If child support sustains the standard of living of children, children would be far less vulnerable because of the poverty that results from divorce. Secondly, Okin believes that the upbringing of children would be less gendered if they have equal input from both their fathers and mothers, if gender roles are less strictly adhered to, and if power imbalances are somewhat rectified. Finally, this kind of family would be a far more effective school of justice than a traditional family. A non-gendered family would better demonstrate fairness and respect, and empathy would be a more universal value rather than just a female one, since doing nurturing would make men more empathetic, and male parents being as present as their female partners will extend the development of the characteristics of empathy and a sense of connectedness that girls experience to boys. Boys and girls would then have the opportunity to become equally good at and interested in both ‘feminine’ nurturing and ‘masculine’ ambition and self-esteem. Okin makes use of Robert E. Goodin’s work on vulnerability to make an argument justifying regulation of marriage on the grounds that it is a particular site of vulnerability. According to Goodin, over and above the usual obligations owed to individuals, ‘we bear special responsibilities for protecting those who are particularly vulnerable to us’ (quoted in Okin, 136). There is also a moral obligation to minimise vulnerability when it is an alterable social artifact because a state of vulnerability

45 Stellenbosch University https://scholar.sun.ac.za opens people up to exploitation. In order to test for whether or not unacceptable, asymmetric vulnerability exists, Okin proposes we apply the test developed by Albert O Hirschman to determine the asymmetry of the relationship by comparing the relative exit strategies of those in relationship: if it is significantly easier for one partner to exit a relationship than another, which thereby gives the former partner significantly more power in a relationship, where the asymmetry is the result of social arrangement, then there exists unacceptable asymmetric distribution of responsibility between them (137). From all the preceding arguments, it should be clear that there is certainly gendered asymmetric vulnerability in heterosexual couples, and this structural reality acts to make it significantly more difficult to leave a relationship as a woman, particularly in the event of housewives. Therefore, even though housewives are often arguably less exploited, they are worse off in terms of their capacity to leave a relationship, and therefore in terms of the corresponding vulnerability.

2.2.3. OKIN AND LIBERALISM

Okin clearly believes that a consistent application of liberalism entails ‘interference’ into the marriage realm. However, this is a controversial statement, given that liberalism is in large part dependent on interfering as little as possible, especially into the private realm (Abbey, 2011:52). As such, Okin could be criticised for insufficiently respecting privacy and protecting individual freedoms. Ruth Abbey lists two potential responses Okin could have to this charge that her interference into privacy is inconsistent with her liberalism (52). Firstly, she could rely on criticisms that feminists have frequently made against the private-public divide, where this divide is merely a strategy for maintaining the status quo, where a refusal to allow the state to regulate the household differently is ‘tantamount to supporting its most powerful members, and thus a violation of the liberal respect for equal personhood’ (Okin quoted in Abbey, 2011:52). Secondly, she could argue that the liberal value of privacy and individual freedom has traditionally only been awarded to the heads of households and has explicitly served ‘to shield family dynamics from scrutiny’ (53). In order that the family is not permanently excluded by default from the realm of justice, Okin can advocate individual privacy, where the family unit (and the attendant relationships between individuals) is not protected from the type of scrutiny which intends to secure universal individual freedom and privacy.

46 Stellenbosch University https://scholar.sun.ac.za

However, the criticism that Okin’s position is importantly inconsistent with liberalism is more serious when one looks at what Abbey considers the most controversial part of Okin’s theory: that the ultimate goal of the regulation of marriage is to do away with gender altogether (Abbey, 2011:54). This is read into her theory from the following hints: firstly, Okin believes that gender structured marriage must be allowed because this is currently what some people prefer, but it must be phased out, and secondly, it will in fact be phased out when these policy changes make an equal division of labour more plausible, and when the bringing up of children does not result in such strict gender roles and gendered expectations. This is partly because of Okin’s non-specified use of gender, but it should be clear from the above that the kind of gender Okin means is the ways in which people are treated (unjustifiably) differently as a result of their gender. Amy Baehr draws a distinction between two kinds of uses of the word ‘gender’ used by Okin, where the first means ‘the institutionalization of sexual difference such that benefits and burdens are distributed according to sex’ and secondly, the ‘psychosexual component of identity’ (Baehr, 1996:54-55). If Okin was prescribing changing the second, this may well seem like an unreasonable (and anti-liberal) interference in individual lives. This resonates with a critique by Robert Card (2001), who believes that Okin’s theory necessitates not only a prescribed change in division of paid and unpaid work in marriages, but also a concomitant change in attitudes (164). His argument is as follows: since Okin relies on Goodin and the extra obligation that arises in cases of vulnerability, any division of labour which causes vulnerability must be prohibited. This is because the major asset that is the consequence of such distribution is itself not distributable: human capital (i.e. earning power). Card argues that the partner who makes sacrifices in her earning power will always have less earning power, which will always be a source of vulnerability for them, and as such no one must be permitted to make such sacrifices: rather, partners must take on exactly the same proportions of paid and unpaid work to mitigate this effect on earning power. In other words, the only equitable distribution of paid and unpaid work in a relationship is exactly equal. Not only does this completely override any preferences partners may have on how to run their home and to do domestic work rather than paid work, Card also believes that it necessitates a change in attitudes, where the change in gender roles must be accompanied by egalitarian beliefs (in other words, about the abolition of gender roles) (167).

47 Stellenbosch University https://scholar.sun.ac.za

I argue that this is a misreading of vulnerability. Firstly, Okin’s policy recommendations do mitigate the vulnerability created even by the disparity in earning power: since the standards of living must be the same in both households, even if the female partner earns less, she must receive enough money to ensure that her standard of living does not suffer as a result. Secondly, I believe Card attributes too much vulnerability to the relative discrepancy in earning power and not enough to the surrounding social conditions which make this into a problem. If the gender roles were reversed, and it is the man that has made more sacrifices in his career in order to take on domestic duties, what would happen to him if he divorced his higher-earning spouse? Firstly, given the current statistics on custody, it is still likely that his wife will be the custodial parent, therefore his household will be smaller and cheaper to maintain. Additionally, he would have far more time to devote to his career than his custodial ex-partner. Secondly, due to the pay gap, he would earn significantly more than a woman in his position. Thirdly, due to gendered expectations, it is more likely that he will have a solid career to fall back on with opportunities of advancement, unlike his female counterpart, who would be more likely to return to a dead-end job. If there is a still a discrepancy between his standard of living and his ex-wife’s, Okin’s policy recommendation are designed to counteract this by prescribing that his and his ex-wife’s standard of living must be the same. As such, I believe that the extra step that Card takes in prescribing a highly specified form of family life is unnecessary. The policy recommendations that Okin makes are designed to protect the vulnerable even if their preferences are detrimental to them, which means that her theory actually protects preferences rather than interfering with them. This counters the objection that she interferes too much in people’s private lives. Okin believes that once society is more egalitarian in terms of gender, these preferences may well fall away and gender will become less and less significant in dictating people’s choices as well as their respective life chances and vulnerabilities. Additionally, it is clear that Okin dislikes gender-structured marriage, but explicitly states that it must be allowed to continue given that it is some people’s preference (1989:180). She thus explicitly protects preference against interference, but ensures that vulnerability is not allowed to continue under the guise of the protection of the private sphere.

48 Stellenbosch University https://scholar.sun.ac.za

2.2.4 APPLYING OKIN’S THEORY TO NURTURING WORK

Okin presents a convincing argument that, firstly, due to the vulnerability brought about by marriage, intervention is justified, and secondly, that such intervention would successfully mitigate such vulnerability. Yet intervention into the division of nurturing work could not be justified by vulnerability. This is because it is not clear that doing more nurturing work makes one more vulnerable – in fact, there is an argument to be made that doing more of it makes one less vulnerable. We saw in the first chapter that the fact that men do less nurturing work in their married lives renders them very vulnerable if they are widowed or separated in older age, since they tend to suffer from severe social isolation, with corresponding health and mental health problems. If we were to apply Okin’s strategy, where people are not made vulnerable regardless of their preferences, then steps would need to be taken in order to ensure that widowed or single elderly men are not isolated – for instance, better assisted care with an eye to addressing social deficits, not merely to addressing physical needs. However, since I believe it is unfair that women do the vast majority of nurturing work, this would not solve the problem: the reason men are so vulnerable upon the loss of their wives is because they depended on them so heavily throughout their lives without learning to reciprocate or to take care of themselves. Men therefore benefit throughout their lives from the distributive arrangement of nurturing work, yet this is precisely what makes them vulnerable in old age. Even if the vulnerability of elderly men is addressed through external supports, the unfair arrangement during marriage is not affected at all. In fact, protecting the vulnerability of elderly men could be seen as facilitating the status quo – which is not to say that the vulnerability should not be mitigated, but doing so will be entirely ineffectual in bringing about a fairer distribution of nurturing work. Even if Okin’s policy recommendations are successful, they will not have a direct effect on the distribution of nurturing work. Okin imagines that the implementation of her policies will have far-reaching consequences on the conceptions of gender: she relies on Nancy Chodorow’s psychoanalytic work that suggests that gendered identities are largely the consequences of having exclusively female caregivers. She suggests that if men are more active parents, they will cultivate empathetic natures, and in modeling this for boys, boys are likely to become more

49 Stellenbosch University https://scholar.sun.ac.za empathetic as well. These are shaky grounds on which to base the solution to the problem of unequally distributed nurturing work: Chodorow’s theory, which attributes so much to having female caregivers, is not in the least uncontroversial. Further, a consequence of Chodorow’s work is that, for example, lesbian parents (who distribute nurturing work much more equally) would produce more ‘gendered’ boys than their egalitarian heterosexual counterparts. There is no evidence of this. Finally, if it is the case that the distribution of nurturing work is unfair to women, then a more urgent intervention is required that is not premised on the dubious possibility of the disappearance of gender altogether after generations of changed policy. To phrase it differently, women should have a way of accessing more nurturing without this being premised on the disappearance of the patriarchy or gender altogether. A woman who is in a relationship with an unfair distribution of nurturing work must have recourse for her particular problem, rather than merely hoping that her future granddaughter does not have a similar problem with her spouse. In other words, it must be possible to claim that the relationship ought to be different, or seek a remedy for its unfairness even while someone is socialised according to his or her gender. Children ought not to be socialised into these harmful gender roles, and adults ought not have to persist in unfair relationships because their partners were socialised as such in their youth.

CONCLUSION

This chapter looked at two contractualist solutions to the problem of the division of household labour and nurturing work. Hampton makes use of a contractualist test that can be employed by an individual to test her relationship for the presence of exploitation. This test may serve some psychological function in encouraging women to cast a self-interested eye over their relationships, but its use is ultimately limited because of the exclusion of benefits and costs that are the side-effects of affective or duty-based ties: as Sample points out, any sacrifice that one makes for one’s family is only explicable in terms of those criteria. Further, since much of nurturing work is intimately tied up to love and duty, their exclusion means that this cannot test for the fairness of the distribution of nurturing work. Its benefit is mainly that it presents a philosophically justified inclusion of self-interest, which could work to counteract the

50 Stellenbosch University https://scholar.sun.ac.za social pressures women feel to be unreasonably self-sacrificing, but it also does not function as a guide for what could constitute being unreasonably self-sacrificing. Okin, on the other hand, makes use of the Rawlsian social contract to imagine a just society that includes a just family. She justifies the seemingly illiberal proposal to regulate family life with the principle that socially constructed vulnerability should be minimised where possible. Her policy-proposals may well be effective in changing the division of housework and thereby the participation of women in the public sphere. Nevertheless, it is unlikely to change the division of nurturing work except insofar as it is likely to change gender altogether by minimising gender roles and gendered expectations. Further, her arguments relating to vulnerability are unlikely to be helpful to the unfair situation women find themselves in in doing more nurturing work, since it is the vulnerability of men that will need to be mitigated. It may be the case that a woman’s emotional well-being is less than her husband’s if she does all the nurturing work, but that need not mean that she is vulnerable, since vulnerability does not simply mean non-optimal conditions, but rather being unreasonably exposed to the threat of harm. The vulnerability of elderly men entails their increased risk of depression, extreme loneliness, illness, and premature death, which means that they are clearly exposed to more risk of harm. The same cannot be said of a woman whose emotional wellbeing is not optimal. Therefore, with respect to nurturing work, Okin’s work is likely to be more beneficial to men than women. Given that I have endorsed her arguments relating to vulnerability, I must also endorse the consequence that the vulnerability of single elderly men ought to be mitigated. Despite this, the unfairness of the distribution of nurturing work cannot be addressed with this theory. These theories cannot tell us why the distribution of nurturing work is unjust: I have shown that it may well maximise utility for a woman to opt into a gender- structured relationship, and can thereby not be avoided by taking rational self-interest into account. Further, since this distribution does not render women vulnerable, it is not clear from Okin’s work that this distribution of nurturing work is an injustice to women. Nevertheless, it is the case that the bulk of nurturing work is more or less automatically allocated according to gender, and so it cannot be merely a case of some relationships being unfair, but rather a way in which women do more unpaid, unrecognised work. As such, I will move onto another candidate for an account of

51 Stellenbosch University https://scholar.sun.ac.za why this distribution of labour could be considered wrongful: an account of exploitation from materialist feminism.

52 Stellenbosch University https://scholar.sun.ac.za

CHAPTER THREE: EXPLOITATION OF NURTURING WORK

INTRODUCTION

The next candidate for the wrongfulness of an unequal division of nurturing work is exploitation. Intuitively, this seems like a useful line of enquiry for the following reasons: 1) we are looking at a kind of work that is not compensated at all, either through reciprocation or through payment; 2) one group of people (those not doing the work) benefit more from this work that is uncompensated (even though it may be considered mutually beneficial), and 3) it seems that this arrangement is in some sense unfair. The standard definition for exploitation is when one party takes unfair advantage of another party at the expense of the latter party, so it seems at its face that this could potentially be a form of exploitation. It is also possible to draw an analogy with a clear case of exploitation, which I take from Ruth Sample’s list of examples:

A factory owner visits a village in a Pacific Rim country and offers to set up a running shoe factory that would pay each worker $2 per day. The current average daily wage in the village is $1, which is enough to prevent a worker and his family from starving. The workers will have no benefits other than salary and must work eighty hours per week. The workers accept. The running shoes sell for $95 per pair in the United States and Western Europe, and half of that price is corporate profit. (2003:8)

The similarities between this scenario and that of the unequal division of nurturing work are as follows: firstly, it is a mutually beneficial arrangement, where the exploited party maximises their utility by entering into it. We saw in the previous chapter that where a woman wants a family, she is maximising utility by having a family, even if she does far more unpaid work than her partner. Secondly, the arrangement is nevertheless unfair, because the exploited party is not receiving her due for the work done. Where the worker in the sweatshop gets paid far too little, employed women in relationships who work do not get compensated at all, either through payment nor through being nurtured in turn. Thirdly, the asymmetry in

53 Stellenbosch University https://scholar.sun.ac.za choices available to the parties causes a power imbalance upon entering into the contract. The sweatshop worker can choose to work for less, but is not able to choose to work for a company that will employ her on fair terms. A man whose wife leaves him because he does too little nurturing work will easily find another woman willing to do that work, whereas a woman is unlikely to find a man who will do an equal share of the nurturing work. He is therefore spoiled for choice for people who will opt in to the agreement on terms that are advantageous to him, whereas she will have trouble finding a man agreeing to such terms at all. Finally, the only route the exploited party has to a certain good is through being exploited: the sweatshop worker can only earn $2 a day by working on unfair terms, and a heterosexual woman can only have a two-parent family if she is exploited. Of course, this is all in very general terms, and it is certainly possible that one out of a hundred workers is employed by someone else on better terms, and that one out of a hundred women finds a man who does half the nurturing work. These exceptions are not enough to render this argument irrelevant. This chapter will look at materialist feminist accounts of how women are exploited by capitalism. In order to do this, an analysis of the mode of production women are engaged in in the family will be provided. This will bolster the materialist feminist point that it is not only the proletariat that is exploited by a capitalist system, but also women insofar as they engage in unpaid work in the home. Secondly, I will look at the work by sociologist Arlie Hochschild about emotional labour in the workplace and its resultant alienation, and attempt to see what parallels can be drawn between the pitfalls of emotion work in the workplace and those in the home. Thirdly, I will look briefly at three additional accounts of exploitation and attempt to assess if they can explain the wrongfulness of an unequal division of nurturing work. Fourthly, I will look at an argument that situates the wrongfulness of an unequal division of nurturing work in the fact that it contributes to the oppression of women.

3.1. ECONOMIC EXPLOITATION OF EMOTION WORK

3.1.1. A MATERIALIST FEMINIST ACCOUNT

Marx’s traditional account of exploitation is based on the industrial mode of production which trades in exchange value. I will briefly outline the traditional

54 Stellenbosch University https://scholar.sun.ac.za account and then explain how materialist feminists extend this analysis to the home. Marx’s account is that of the relationship between the worker and the capitalist, who owns the means of production. Workers do not own the means of production, and so are forced to sell their labour power in order to survive (Colley, 2015:223). Workers believe they are selling their labour, which means that they think there is a one-to-one correspondence between the work done and the pay received, and so are fairly compensated for their labour. However, this belief forms part of capitalist ideology that functions to prevent workers from realising that they are being exploited. In fact, workers sell their labour power, which allows the capitalist to extract as much labour as possible from the worker with the same amount of pay (224). The worker thinks he is being paid per hour of work, and that this is a fixed unit of work, but the capitalist tries to extract as much labour as possible from this hour of paid work. The wages that workers receive, rather than being determined by the proportion of the value of the object, is determined in ‘the relation to the costs of ensuring that the labourers can socially reproduce themselves and continually supply labour power’ (224). In reality, their work is worth much more than they are paid, and the proportion of their labour which they are not paid for is known as ‘surplus value’ and is appropriated by the capitalist and called ‘profit’. This value is extracted from the gap between the pay that the labourer’s family needs to survive and the real value of the product or commodity produced. Marx explains how labour power reproduces itself: workers sell their labour power, and then they use the money earned to sustain their lives, and as such, reproduces labour power. However, the worker still has amassed no capital, and so must work again in order to eat and survive (Himmelweit, 1991:200). However, according to Himmelweit, Marx did not give any account of how workers are reproduced, where one generation of workers is replaced by the next (202), and as such makes the proletariat and the capitalist ‘appear as immortal and therefore not in need of reproduction themselves’ (202). This oversight is particularly relevant for the working class, since the capitalist class need not consist of any great number, and in fact how many people occupy this class does not affect the functioning of the economic system. However, the proletariat and the labour power they sell consist of living bodies, and their role ‘depends on their physical reproduction as human beings’ (210).

55 Stellenbosch University https://scholar.sun.ac.za

Feminists have attempted to address this lacuna in Marx by extending the analysis of labour relations to the family. One such feminist is Christine Delphy, who, in her book, Close to Home (1984), distinguishes between the ‘industrial mode of production’ and the ‘family mode of production’. The key difference usually asserted between the labour in the industrial mode of production versus the labour in the family mode of production is that whereas the industrial mode of production produces value that is sold for consumption by others – which is known as exchange value –in the family mode of production, the value produced is consumed within the family without being sold, which is known as use value. Exchange value is usually considered productive labour and use value is typically considered reproductive labour. However, this very distinction reveals a masculinist bias in the way that labour is theorised (Colley, 2015:231). Separating labour into ‘productive’ and ‘reproductive’ is a way of placing men’s work above women’s work in a hierarchy, and is thereby ‘ideologically entangled with capitalism’s undervaluing of women’s work’. Colley argues that capitalism works to obscure women’s work by making it seem that the industrial mode of production is a relation between the worker and his employer, when in actual fact it is a three-way relationship between the female doing nurturing work, her husband working for a wage, and the capitalist paying a wage, where the male worker only acts as a middleman between the woman doing ‘reproductive labour’ (in other words, creating labour power), and the capitalist appropriating that labour power by underpaying the employee (231). Materialist feminists therefore argue that capitalism produces an ideology that makes it appear that women’s work is a matter of ‘natural’ proclivity, of personal relationships, of use-value (self-consumption), and thereby as non-work, whereas it is actually integral to capitalism’s functioning and the industrial mode of production, and insofar as Marxian theories of labour exclude women’s work, it is complicit in capitalism’s ideological undervaluing of women’s work’ (231). Delphy argues that the distinction between use-value and exchange value is not at all clear-cut. Family work is traditionally considered use-value rather than exchange-value because it is consumed by the family rather than exchanged for money. However, Delphy points out that an agricultural family challenges this strict distinction: in agricultural homes, women are often responsible for things that can both be consumed by the family and sold in the market, such as being in charge of the

56 Stellenbosch University https://scholar.sun.ac.za hens and the milking. The family both consumes and sells eggs and milk, which itself challenges the notion that use-value (in other words, what is intended for consumption), and exchange-value, are distinct (61-62). If we were to look beyond the agricultural family and look at the family that earns money, the kinds of domestic services performed by women can similarly be both consumed by the family, and sold on a market, and when women do this work outside of the family, they are paid as wage-labourers, for instance as cleaners or child-minders. Domestic services are thus clearly not the kind of services that produce only use-value intended for self- consumption. Delphy suggests that for these reasons, a woman’s labour power is appropriated by her husband in the home, which is to say that it is (and its fruits are) unfairly taken possession of by her husband: ‘in family production, she does not make use of her labour power; her husband makes use of it’. Insofar as the domestic services a woman provides sustain life that earns money on the market, her husband trades what she provides for money that he receives for his work (67). The modes of production the husband participates in are distinct from the modes of production his wife participates in (in her unpaid labour). The industrial mode of production is characterised by the following: the services provided by the worker are fixed in nature and quantity, as is the value attached to those services (70). In contrast, a woman who works in the family mode of production performs services which are not fixed in their nature, nor is any fixed value accorded to them. Rather, both the content of the services and the value thereof depend on the inclinations, needs, and wealth of her husband. The wife of a wealthy man can earn significantly more for the same services than the wife of a poor man, since what qualifies as maintenance for the wife will depend on her husband’s standard of living. Further, the industrial mode of production is characterised by the fact the individuals participating are theoretically interchangeable, so the employee can be replaced and the worker is free to offer his services to a different employer. Workers are typically also able to improve their standard of living by working harder or providing higher quality work. In the family mode of production, on the other hand, the family relationship is highly individualised, and so wives are not able to offer their services to a new ‘employer’. They further cannot improve their standard of living by working more and improving their services, since their standard of living is entirely dependent on that of her husband (70).

57 Stellenbosch University https://scholar.sun.ac.za

The housewife, or the wife insofar as she performs domestic services, thus falls outside of the proletariat/capitalist class system, since she does not actually belong to the same class as her husband. Her husband owns the means of production and appropriates her labour power. Delphy suggests that this economic relationship most closely resembles serfdom, which is to say that it is labour power exchanged for maintenance, and wives constitute their own class distinct from their husband. Because women are destined from birth to become a part of this class (since nearly all women marry), women constitute a caste (71). This analysis is a technical one rather than normative, but it is trivial to state that the existence of a caste (where one is destined to be disadvantaged because of an arbitrary and unchangeable factor) is a matter of injustice. Further, since women’s labour power is appropriated and they do not own the means of production, they are exploited. I have until now maintained that housewives are less exploited than working wives because they at least receive some financial compensation for their housework, whereas working wives do the work without that benefit. However, this current economic analysis calls that view into question: since neither the housewife’s services nor their value are fixed, it is not at all clear that she is being paid for that work. Rather, she works in exchange for her keep, and the standard of living she receives is entirely dependent on her husband’s class status instead of her own work. This is also true of working wives, except she has an income insofar as she is active in the industrial mode of production, but she is not guaranteed maintenance. In fact, Delphy states that the costs of maintaining the home and children typically comes from the wife’s salary, so she pays for the maintenance of herself and her children, and possibly part of the maintenance of her husband. This means, argues Delphy, that it is ‘as though she never earned the money in the first place’ (69). This economic account of exploitation shows that it is not only the proletariat that is exploited, but also the women who reproduce the proletariat. Capitalism exploits women in a particular way, and so the gendered oppression of women in heterosexual relations is intimately tied up with capitalism. Insofar as women perform more unpaid labour in the home, they are exploited in this way, and since emotion work constitutes a part of such reproductive labour, it is exploited. This, however, chiefly serves as a feminist critique of capitalism, rather than a critique of the way people structure their family lives. However, the critique that workers are alienated from their labour is a normative claim about how the labour relation is bad. I will thus

58 Stellenbosch University https://scholar.sun.ac.za pursue how emotional labour in the workplace can lead to alienation, using Hochschild, and then attempt to extend this analysis to the home.

3.1.2 ALIENATION OF EMOTIONAL LABOUR

We saw in the first chapter that Arlie Hochschild offers a new account of emotions which does not describe emotions as a spontaneous, uncontrollable response to events, but rather is the product of complex rules and effort. We frequently attempt to manage our emotions in such a way that they obey ‘feeling rules’. We also do emotion work on others, where beyond the work we do on our own emotions, we target the emotions of others. The ways in which we do this on ourselves and on others in our private lives in order to improve the emotional or mental states of others, or the wellbeing of the group, is known as ‘emotion work’. In her book, The Managed Heart (2003b), Hochschild is interested in the way in which this private phenomenon has been utilised in the realm of business, where it becomes a use-value that generates profit. She refers to this phenomenon as the ‘transmutation of an emotional system’, and she refers to the emotion work that occurs in the workplace as ‘emotional labour’ (19). Hochschild’s book is chiefly an empirical study of flight attendants as an example of performing emotional labour. Specifically, she looks at what happens in the case of a ‘speed-up’, where planes become larger but crews of flight attendants become smaller, which means that every individual flight attendant has more passengers to attend to (122). In this case, it becomes impossible to invest emotional labour with the sincerity that comes with ‘deep acting’, which involves generating the appropriate emotions, as opposed to ‘surface acting’, which is merely pretending to be feeling the appropriate emotions. The flight attendant is therefore unable to generate genuine friendliness and concern for the passengers’ wellbeing. In some cases, the performance of emotional labour simply becomes impossible – for instance, a woman explained how she avoids eye contact because passengers typically use eye contact to make a request, and she has no time to attend to requests (122), and so she avoids eye contact altogether. As such, these flight attendants, ‘when asked to make personal human contact at an inhuman speed, they cut back on their emotion work and grew

59 Stellenbosch University https://scholar.sun.ac.za detached’ (126). This is known as ‘going robot’, where the flight attendants adopt surface acting rather than deep acting. The impossibility of engaging in deep acting has three potential effects: the first is burnout (where the worker finds herself cut off from her emotions in the form of ‘emotional numbness’); the second is a feeling of inauthenticity; and the third is cynical acting (where the worker becomes estranged from the acting itself) (188). The speed-up need not be understood in terms of surplus value, but can still take a contribution from Marx in that the speed-up and its concomitant problems are clearly the result of conflicting interests between employer and employee: the employee desires to take pride in her work which she does well, which means that she needs to be able to engage in deep acting rather than surface acting, which she is only able to do if she has enough time and enough resources. Conversely, the employer needs her to ‘process’ as many people as time allows, not as emotional resources allow. This conflict caused by exploitation results in harm that the employee suffers, and this harm occurs chiefly in her sense of authenticity. In his paper, ‘The Alienated Heart’ (2009), Paul Brook examines what he calls ‘Hochschild’s half-made theory of alienation’ (14). He dubs it thus because Hochschild only gives an account of two of the four elements of alienation as identified by Marx: she gives an account of product alienation and process alienation, but not of alienation from fellow beings and alienation from human nature (9). Flight attendants are alienated from the product of their labour because the product of their emotional labour belongs to the company (which is sold to the consumer) rather than to them. Their friendliness, their smiles, and their solicitousness are all seen to belong to the airline, and are part of the product that the customer pays for, instead of belonging to the individual flight attendant. The consequence of this is emotive dissonance as described above, where the flight attendant experiences burnout, cynicism, or inauthenticity. According to Brook, emotive dissonance is ‘a more profound form of alienation than even Marx imagined’ (16). A flight attendant is alienated from the labour process because, in the ‘speed-up’, her work is de-skilled by the codification of her work (17). This occurs in the form of strict rules flight attendants must follow, such as being given a script when greeting passengers or offering them beverages, which means that she cannot use her own initiative to use her emotional labour effectively. Freedom, discretion, and personal creativity are thus all stifled in the process, again working against deep acting. Brook argues that though

60 Stellenbosch University https://scholar.sun.ac.za these are the only two kinds of alienation Hochschild looks at explicitly, she ‘leaves open the theoretical door’ to the examination of other kinds of alienation that occur in the commodification of emotional labour (21). This will be not be expanded upon here, since it is chiefly of interest whether or not Hochschild’s analysis could plausibly be translated to emotion work in the domestic space, and whether these two kinds of alienation Brook identifies occur in the family. The question is now whether or not the emotion work done at home is of the kind of nature that it can be exploited, and therefore if it can lead to alienation. In order to do this, we need to point out the relevant similarities and differences between emotion work and emotional labour, that is, the unpaid variety in the intimate context of the family, and the commercialised version that occurs in the workplace. These are similar in that they are both subject to ‘feeling rules’, though one might be literally formalised (in the form of scripts and the like) while the other remains informal, and they both preferably require ‘deep acting’ rather than surface acting, where one does not simply pretend that one is feeling feelings, but rather attempts to actually feel them. In both cases, successful emotion work/labour must be performed in such a way that the effort involved must be as invisible as possible – in other words, it must seem as sincere as possible. The most important difference that needs to be pointed out is that, whereas, for example, the flight attendant is wearing the smile that represents her company, a woman in the family context wears simply her own smile. In other words, the goals she seeks to achieve are her own rather than that of a company. Further, the process is not codified at all. A woman in the private sphere may typically freely and authentically choose how to do emotion work. A second salient difference is that the workplace is subject to a ‘speed-up’, where a worker’s working conditions can be changed in such a way that it becomes impossible to do her emotional labour with the sincerity that comes with ‘deep acting’, whereas in the home this is unlikely, as discussed below. However, it is worth complicating these differences somewhat, starting with the latter. The example of speed-up Hochschild uses occurs among flight attendants, where a flight attendant has to engage with 155 people in 40 minutes. It is of course nearly impossible that a woman would face this kind of pressure in her intimate relationships. However, if she is the kind of woman who has a full-time job, small children, and takes on the second shift, her resources may well be taxed to the point

61 Stellenbosch University https://scholar.sun.ac.za where she is not able to invest the required deep acting into her emotion work. If she finds it important to keep doing emotion work regardless of her inability to achieve sincerity in doing so, it seems possible that she may show one of the three expressions of alienation Hochschild identifies: burnout, feeling ‘phoney’, or becoming cynical in her ‘acting’. Therefore, even if the ‘product’ of her work is part of her own goals and desires (rather than belonging to a company), and even if she is almost entirely in control of how she chooses to go about doing this work, it is nevertheless possible that she bears a cost in terms of her sense of authenticity. It is not implausible that some women would end up in this situation, particularly those who feel the expectation to achieve ‘supermom’ status. Nevertheless, it is also possible, and perhaps somewhat more likely, that a woman might simply opt out of doing this emotion work in the home, where she refuses the pressure to put work into conversational cheerleading, to feeding egos, and tending wounds. This possibility to opt out, where a woman might choose to maintain her authenticity instead of risking insincerity, is a significant difference between emotional labour and emotion work. This is not to say that there is no cost: the harmony of the family and the happiness of the family members, as well as the health of the relationships, will certainly suffer from the absence of such work. Given that women are socialised into associating the successful performance of emotion work with their identities, opting out of this work may also have consequences for her self-esteem and happiness. Nevertheless, these (unattractive) options are still available. Therefore, in some situations, there is sometimes harm attached to women being expected to do the bulk of emotion work, and that cost is alienation from the self. From a utilitarian perspective, there is perhaps a reason to think that this is why the unequal distribution of labour is wrong. However, it is certainly only in some circumstances that this particular harm occurs – it is only when a woman is entirely overburdened, which may well be only for the years when her children are young. This would mean that, once her children are older and her workload decreases slightly, it would be acceptable for her to still do the bulk of emotion work in her household, but this is clearly still wrongful even if it no longer impacts her authenticity or leads to burnout. This argument, which ultimately rests on a utilitarian account, can therefore not explain what is wrong about the unequal distribution of nurturing work in the home.

62 Stellenbosch University https://scholar.sun.ac.za

There is a further reason for thinking that the harm of the distribution of labour is not sufficient to show the wrongness of exploitation in this case, which can be found in Ruth Sample’s critique of Jean Hampton in Chapter 2.1. To recount it briefly, Sample analyses the decision of a woman to stay in a relationship from a rationally self-interested perspective. She concludes that if a woman wants a family and children, it is likely that she maximises her utility by doing so, even if she does so under unfair conditions. Therefore, even if women experience harm by doing more emotional work, they are voluntarily present in a relationship and can choose the alternative. The benefits thus outweigh the harm, and utilitarianism cannot explain why the distribution of labour is wrong. We thus need to draw on a notion of exploitation that tries to determine what is unfair about someone doing more work than they are compensated for.

3.2. ALTERNATIVE ACCOUNTS OF EXPLOITATION

In her book, Exploitation and Why it’s Wrong (2003), Ruth Sample looks at two alternative accounts of exploitation in order to formulate a new definition of exploitation that avoids the pitfalls of these two. These alternative definitions of exploitation also produce different answers to the question of why exploitation is wrong. I will briefly explain each, and assess whether the unequal division of nurturing work qualifies as exploitation under that definition, and if it can explain its wrongfulness.

3.2.1. WERTHEIMER’S ‘MARKET PRICE’ EXPLOITATION

Sample explains that Alan Wertheimer’s account of exploitation is ultimately outcomes-based rather than focused on procedural unfairness (Sample, 2003:16). His account is a conventionalist one, which is to say that his definition of exploitation depends on whether or not an existing convention is observed. The convention he utilises is market price: if someone is paid less than a hypothetical market price (in other words, the price that is determined by the supply and demand of the market), then they are being exploited, and if someone is paying less than market value, they are exploiting someone (19). However, Wertheimer acknowledges that exploitation

63 Stellenbosch University https://scholar.sun.ac.za does not only occur in business relationships, but also in intimate relationships, and here he invokes a ‘baseline of decency’ (23). But Sample argues that the ‘market price’ test actually does no theoretical work; rather, falling below the baseline of decency is always required for something to constitute exploitation. For example, some cases of exploitation are exploitation because the market price is too low: for instance, in the case where someone in a sweatshop is paid $1 for every shoe they produce which is then sold for $98, this is considered by many to be exploitative, particularly if that income is not enough to sustain a decent life. However, it is because of the excess of supply of workers and the comparatively low demand of the capitalist (in other words, there is one capitalist providing jobs and many workers competing for those jobs) that the market price for labour is so low. The reason why this is exploitative must then be because it falls below a baseline of decency. There is no account provided for how to determine a ‘baseline of decency’, so Sample concludes that this must also be determined by a hypothetical market price. For instance, if we have a husband and wife, and the wife makes more sacrifices in her career in order to take on more unpaid work for her family. As indicated earlier, the reason why this is so prevalent must be because of something akin to a hypothetical market price: there is a highly limited supply of gender-egalitarian men who are willing to make an equal amount of sacrifices in the workplace and do extra work at home. Conversely, there is a large supply of women who are willing to do this. We saw in the previous chapter that if a woman wants a partner and children, it would maximise her utility to participate in family life on exploitative terms rather than be single, childless, and not exploited. However, the maximum utility a man can achieve is higher than that of a woman: he need not choose between being exploited or childless; he can both avoid doing an unfair amount of work and have children. He can have a fulfilling, successful career as well as a family. The ‘market’ is thus unfair to begin with, where one group of people benefits from the status quo. Even though Wertheimer acknowledges that even mutually beneficial arrangements can be exploitative, he cannot account for the badness of exploitation where the status quo already exhibits an injustice. His theory of exploitation is therefore not useful in explaining the wrongfulness of an unequal distribution of nurturing work.

3.2.2. GOODIN’S ACCOUNT OF EXPLOITING THE VULNERABLE

64 Stellenbosch University https://scholar.sun.ac.za

A further account of exploitation that Sample looks at is that of Robert E. Goodin, whose work on vulnerability Susan Moller Okin uses, which appeared in Chapter Two. In his chapter, ‘Exploiting a situation and exploiting a person’ (1987), Goodin defines exploitation as ‘taking unfair advantage of someone’, which is a moralised definition of exploitation, which means to say that x is a case of exploitation is to say that x is morally wrong. The wrongness of exploitation is thus in its very definition. He frames ‘unfairness’ in terms of the analogy of games: ‘the rules of the game’, where ‘playing’ fairly means obeying the rules, and ‘playing’ unfairly means breaking the rules (1987:183). Different contexts and interactions have different rules of engagement (rules of the game), and deviations from this would depend on those rules of engagement. The rules of engagement are simply those that are stipulated by existing norms, which means that Goodin’s account is also a conventionalist account. It would thus be exploitative to gain an advantage over someone by violating the existing norms of the game. This is the first way of exploiting someone. The second way of exploiting someone is when one does play according to the rules of the game, but the party one is engaging with is vulnerable, which makes it inappropriate to play for advantage (184). Goodin lists four kinds of situations where the other party’s vulnerability makes it inappropriate to play for advantage: i) when the person one is playing against has herself renounced playing for advantage by, for example, acting based on trust; ii) when the player is unfit to play for advantage; iii) when players are no match for you in games of advantage; and iv) when your relative advantage derives from others’ grave misfortunes. In each of these cases, the party one is engaging with is vulnerable, and that vulnerability makes it wrong to play for advantage (185-186). Presumably, when Goodin says the ordinary rules of engagement ought to be suspended, he does not mean that behaviour that would normally be considered unfair would be allowed, but rather that the other person’s interests deserve especial consideration. This corresponds to Goodin’s argument in his book, Protecting the Vulnerable (1985), where he argues that we have a special obligation to protect the vulnerable. A minimum requirement of this is to refrain from exploiting the vulnerable, which means refraining from playing for advantage against them.

65 Stellenbosch University https://scholar.sun.ac.za

The first case of vulnerability, where the other person has renounced playing for advantage, seems promising, since intimate relationships are characterised by love and trust and are usually not expected to consist of two people attempting to extract maximum personal advantage from each other (or that this would be occurring unilaterally). However, since this is the case, it seems highly unlikely that the extremely prevalent unequal distribution of nurturing work between men and women is because men are playing for advantage against women who have renounced doing so. Rather, the men in these relationships most likely see themselves as having renounced playing for advantage as well. I will return to this point in a moment after briefly dispensing with the other kinds of vulnerability. The second, where the other player is unfit to play for advantage, also would not apply in most cases. In fact, we saw in Sample’s critique of Okin in section 2.1.3. that individual women may well be maximising utility by entering into these relationships, even if the terms of such relationships do not provide equal utility to the male and female partners. A player would be unfit to play for advantage due to some attribute they have: perhaps they are unconscious, or too young, or not compos mentis. It is certainly not the case that most women are unable, in principle, to play for advantage. The third, when the player is no match for you in terms of advantage, seems to refer to massive power differentials – Goodin uses the example of someone having a monopoly on something that others need desperately. In this scenario, those with the desperate need do not have sufficient bargaining power because of their situation. This would also not apply to our scenario. The fourth case is where one’s relative advantage is the consequence of someone else’s grave misfortune. This one is interesting, because while it is the case that the relative privilege, status, and resources men possess are absolutely dependent on the oppression of women, it is still not clear that the social lot of women could be characterised as ‘grave misfortune’. In fact, it seems that in order for something to be considered a ‘grave misfortune’, it would need to render the party extremely vulnerable. I discussed the application of Goodin’s work in Protecting the Vulnerable to this phenomenon by Okin in section 2.2.3, and showed women are rarely, if ever, made vulnerable by the unequal distribution of nurturing work, and elderly, widowed men are made the most vulnerable. As such, this fourth scenario would also not apply. There is an element of Goodin’s theory aside from vulnerability which makes it difficult to apply to intimate relationships: formulating it as ‘playing for advantage’

66 Stellenbosch University https://scholar.sun.ac.za suggests deliberate activity, as consciously attempting to maximise one’s benefit from an interaction. This formulation of course makes sense with business interactions, where business negotiations are explicitly characterised as each party attempting to extract maximum advantage, but it makes less sense with personal relationships. Sample points out in her own account of exploitation that it occurs because men and women have gendered expectations of how much work they can expect their partner to do, and because they are correct in those expectations (2002:266-267). The fact that we have been conditioned into gendered expectations is the root of the exploitation, not the deliberate actions of men in relationships. This is not to say that there is no personal moral obligation to refrain from exploiting another in an intimate relationship, but it does mean that deliberately ‘playing for advantage’ is an incorrect way of characterising the phenomenon. If it is the case that women are not necessarily made vulnerable by the unequal distribution of nurturing work, then all that remains of Goodin’s theory to make an argument for exploitation is his standard formulation, which would be taking unfair advantage of another person, where unfairness refers to a violation of the social norms governing the interaction. This will also not work with this phenomenon because the norms governing heterosexual relationships are precisely the problem, and so the same problems as with Wertheimer’s theory of exploitation reemerge in Goodin’s. Given that women are taught to express love through doing nurturing work in a way that men are not, the very conception of romantic love between heterosexual partners currently expects women to do more nurturing work than their partners, whereas male partners are expected to contribute more financially. Relying on convention as a measure of fairness will never detect the unfairness that arises from an existing societal oppression. Until our very notion of romantic relationships, and what constitutes fair play in such relationships, encapsulates an idea of either an equal division of nurturing work, or at least that such work should not be distributed automatically on the basis of gender, a theory of exploitation relying on convention cannot be used to show that the distribution of nurturing work is exploitative. Interestingly, Goodin’s formulation could be used to show that an unequal distribution of nurturing work where the male partner does more nurturing work is exploitative, and this is precisely because it is an unconventional distribution of labour. Romantic relationships are regulated by norms that allow for women to do more nurturing work, but not for men, and insofar as a woman gains an advantage

67 Stellenbosch University https://scholar.sun.ac.za from her male partner doing more emotion work, and insofar as that deviates from the norm of a romantic relationship, she exploits her partner. In other words, the case of Tom and Emma in the introduction could be declared wrongful because it deviates from gender roles. However, this only serves to highlight the problem of a conventional account of exploitation, since the wrongfulness of that arrangement is certainly not that it deviates from the usual sexist patterns. Therefore, Goodin’s theory of exploitation cannot show the wrongfulness of an unequal distribution of nurturing work in a romantic relationship that is such because of allocation based on gender or a gender atypical allocation.

3.2.3. SAMPLE’S ‘EXPLOITATION AS DEGRADATION’

Ruth Sample puts forward an account of ‘exploitation as degradation’ (2003:56), where she posits that exploitation occurs when one degrades or disrespects someone in order to gain an advantage (57). Sample divides disrespect of persons into three categories: i) failing to provide what is necessary for someone’s flourishing, ii) taking advantage of an existing injustice, and iii) commodifying something about a person that ‘ought not be commodified’ (57). Some of the advantages of this view according to Sample are that degradation is already considered wrong, and so her account has a clear normative element (59). Secondly, it is possible to account for exploitation on both an individual and institutional level – unlike, for instance, Wertheimer’s view, this account does not rely on convention, which may already enshrine an injustice (60). A further advantage of this account is that one needn’t include intent or even awareness that exploitation is taking place: it is possible to exploit by degrading someone without intending to do so or without knowing that one is doing so. Further, it can be used to identify the presence of exploitation even in a mutually beneficial relationship. In other words, even where there is no net loss in utility from the perspective of the person being exploited, it still qualifies as exploitation. It now needs to be established whether or not this theory sufficiently accounts for the exploitation that ostensibly occurs in gendered nurturing work. In order to do so, it should be possible to show that, firstly, men in these relationships gain an advantage through unjust arrangements, and secondly, that women are degraded in

68 Stellenbosch University https://scholar.sun.ac.za one of the three ways through this unequal (or unreciprocated) division of nurturing work. Firstly, this is not a case where someone is being deprived of something they need for their wellbeing or flourishing through neglect, and so the first kind of degradation is not at stake here. While it is true that women are deprived of care specifically from their partners, which may well be detrimental to the wellbeing of that relationship, it is in principle possible for a woman to access care and intimacy elsewhere – perhaps from her female friends or family members. It is not the case that the man with whom a woman is in a relationship with has a monopoly on the goods that a woman needs to achieve wellbeing. It seems possible that a woman could be in an extremely unequal relationship with respect to nurturing work and nevertheless flourish because of being nurtured by her other relationships. This is not to say that one’s partner does not have some kind of monopoly (see Chapter 4), but if one’s partner withholds nurturing work, one still potentially has access to nurturing work from other people. Likewise, the third kind of degradation that takes the form of commodifying something that ought not be commodified is unlikely to be useful. It is not clear that emotion work ‘ought’ not be commodified. Perhaps it is the case that there are things in an intimate relationship where reciprocation cannot take the form of money, which will be discussed at length in Chapter 4, but this is a different argument. While it seems correct to say that women should not have to pay off their financial ‘debt’ to their husbands by doing extra emotion work, it does not seem correct to say that it is not the kind of work one ought to be paid for. In Hochschild’s account of the flight attendants, it seems that the problem occurs when conditions are present to cause alienation. This suggests that there are conditions where salaried emotional labour can be meaningful, fulfilling work. It would be controversial to state that it is wrong of a company to require their employees to be genuinely friendly or to seem interested in the concerns of their customers. In other words, the mere fact that someone is paid to perform emotional labour is not sufficient to call that work exploitative. Further, even if it was the case that emotional labour ought not be commodified, it is not clear that commodification of emotion work is what is occurring in intimate relationships. Even women who earn more than their husbands and therefore do not have a financial debt to pay off do more nurturing work than their husbands. The second form of degradation is the taking advantage of an existing injustice. This division of nurturing work is the result of patriarchy, where women are

69 Stellenbosch University https://scholar.sun.ac.za expected to do more of this work than men are, and women face more serious costs for opting out of this kind of work than men do: we saw in Chapter One that women are held more accountable to the performance of emotion work than men are. In that way, then, many cases of such an unequal division of labour in personal relationships are taking advantage of an existing background injustice. This definition of exploitation is thus promising: women are degraded by the unequal division of nurturing work because it entails men taking advantage of an existing injustice, and this constitutes exploitation. It is probably also the case that this gendered division of labour would be far less prevalent if this injustice did not exist – if men faced the same costs for not doing nurturing work that women currently do, and if this constituted a motivation to do more nurturing work, then men would probably do more emotion work in a relationship. I still have a concern with this definition of exploitation, however. I want to provide an account of the wrongfulness of the unequal division of emotion work that is not wholly dependent on the oppression of women. We can imagine that a relationship in the present day (with its concomitant gender injustice) where a man does nearly all the nurturing work and the woman does almost none might still constitute an unfair distribution of nurturing work. In other words, the wrongfulness of an unequal division of labour is not merely that it is gendered – it is wrong in and of itself. This needs to be argued, and I will do so in the next chapter, but will simply note here that this is why I do not accept at this point that Sample’s ‘exploitation as degradation’ is the full explanation for the wrongfulness of an unequal division of nurturing work. However, in the next sub-section I look at precisely this kind of feminist argument in more detail, in order to flesh out this claim that an unequal distribution of nurturing work contributes to gender injustice.

3.3. EXPLOITATION AS CONTRIBUTING TO THE OPPRESSION OF WOMEN

There is a third candidate for the wrongfulness of exploitation, and that is that it contributes to the oppression of women. In her book, Femininity and Domination (1990), Susan Bartky presents two ways in which it does this. Firstly, she argues that doing emotion work is a way of according status, and secondly, she argues that it contributes to the epistemological oppression and exclusion of women. I will tie this latter argument to Miranda Fricker’s book, Epistemic Injustice (2007), where she

70 Stellenbosch University https://scholar.sun.ac.za provides an account of ‘hermeneutical marginalisation’, and I will argue that unequal distribution of emotion work contributes to the hermeneutical marginalisation of women. I will begin by explaining Fricker’s work. Briefly, ‘hermeneutical injustice’ refers to wrongful lacunae in hermeneutical7 resources, and ‘hermeneutical marginalisation’ refers to the systematic exclusion of a group of people from the production of hermeneutical resources. Hermeneutical resources are those epistemological tools we use to make sense of our experiences. These phenomena can be explained as follows: society is structured by power relations, and it is the case that those who are more powerful have an unfair advantage in the formation of social meanings (Fricker, 2007:147), including in the production of hermeneutical resources. This is, in itself, not a problem for the majority of resources created, since both the oppressors and the oppressed in this case are humans, and as such have a large overlap in their experiences. However, if there is an experience that is specific to the oppressed group, perhaps as a result of their oppression, it is probable that there will be no existing hermeneutical resources to make sense of those experiences. Where there is an experience but no resources to make sense of that experience, there is what is known as a ‘hermeneutical lacuna’ (150). A good example of this phenomenon is that of sexual harassment. Before there was a name for sexual harassment, ‘there was a lacuna where the name of a distinctive social experience should be’ (151). There were individual instances of being touched inappropriately or sexually propositioned that could be laughed off by men as merely ‘flirting’ or ‘joking’. However, women experienced cognitive dissonance between the words of men and their own experiences: men called it flirting, but they experienced it as uncomfortable, degrading, and sometimes even traumatic. This example also illustrates what is at stake with hermeneutical marginalisation: women who quit their jobs due to sexual harassment could not explain why they had done so, and so missed on unemployment benefits, and advocacy to prevent sexual harassment in the workplace could only occur once it had already been labeled (150).

7 ‘Hermeneutics’ refers to the interpretation of texts, as well as the interpretation of the ‘whole social, historical, and psychological world’ (Blackburn, 2005: 165). Fricker’s work is of course about the second kind of hermeneutics rather than the first.

71 Stellenbosch University https://scholar.sun.ac.za

Susan Bartky, in her essay, Feeding Egos and Tending Wounds (1990), draws an interesting link between an unequal distribution of emotion work and the hermeneutical marginalisation of women.8 Bartky argues that the wrongfulness of such an unequal distribution of emotion work lies in the fact that it contributes to the oppression of women. One of the ways in which it does this is that it contributes to hermeneutical marginalisation. Her argument goes as follows: it is implicit in caring work that one has an ‘epistemic lean’ towards the perspective of the care-receiver (1990:111). This ‘epistemic lean’ is the opposite of a hermeneutic of suspicion: in order to do emotion work, one must believe the account which is being presented, and respond with sympathy to the way the speaker is situated in that account. This could involve, for example, responding with indignation to the reprimand of an employer, instead of pointing out that the reprimand may have been justified, or taking for granted the truth of the account instead of wondering if the harms committed have not been exaggerated (Bartky, 1990:111). As such, caregiving work often involves seeing the world from someone else’s perspective, ‘seeing the world according to him’ (112). Since it is the case that women do more of this specific form of caregiving than men do, this implies that seeing the world ‘according to him’ happens far more often than seeing the ‘world according to her’. As such, the hermeneutical resources he uses to make sense of his experiences are bolstered by her sympathy, whereas hers never are. The nurturing work expected from women thus perpetuates that status quo that hermeneutical resources generally and also more broadly culturally favour the experience of men. She argues that a further problem with an unequal division of emotion work is that it involves an asymmetrical bestowal of status. She draws on the sociologist Theodore Kemper, who argues that, rather than romantic relationships being a site where status is irrelevant, ‘a love relationship is one in which at least one actor gives (or is prepared to give) extremely high status to another actor’ (Kemper quoted in Bartky, 109). The kind of emotion work studied by Hochschild that flight attendants

8 Bartky’s work precedes Fricker’s by some decades, and so she does not use the terms ‘hermeneutical injustice’ or ‘hermeneutical marginalisation’, but this essay of hers is certainly about the same phenomenon, so I will keep using this term for convenience.

72 Stellenbosch University https://scholar.sun.ac.za typically perform is exactly the kind that elevates status: it is work that functions to make the interlocutor feel important and heard, and to give him a platform on which to express his views and feelings. Emotion work thus constitutes a kind of status accord, where the self-esteem of individuals depends on how much emotion work is done for them. When this caregiving is asymmetrical, it constitutes and reinforces a performance of higher status for the man and a lower status for the woman. Whenever he does not reciprocate her caring work, ‘he confirms for her, and just as importantly, for himself, her inferior position in the hierarchy of gender’ (109). For Bartky, asymmetrical nurturing work is no less than a perpetual affirmation of the superiority of men and the inferiority of women, and as such, perpetuates patriarchy in its performance. This is therefore another way in which disproportional emotion work contributes to the oppression of women. Let us consider the cumulative effect of gendered status accord. If the distribution of emotion work is gendered in the majority of heterosexual couples, this means that effectively most of the population is taught that women are inferior on a daily basis. A woman who performs emotion work without reciprocation does not only demonstrate her male partner’s superiority to herself and to him, but also to her children. It is plausible that this contributes to the lower status attributed to women by society, and thereby contributes to the oppression of women. Despite the fact that Bartky does not believe that Marxism applies successfully to emotion work in the home, she nevertheless draws an interesting Marxist conclusion: even if work in the home does not match a Marxist account of labour and the exploitation thereof, there is a way in which hermeneutical marginalisation causes alienation. She draws on Marx’s account of the human being, where we are differentiated from animals not just by our productivity, but because we create cultural artifacts and meanings, including institutions, belief systems, and sets of assumptions (1990:34). She refers to the exclusion of women from participating in the generation of social meanings as ‘the cultural domination of women’ (35), which is the same phenomenon as Fricker’s epistemic and hermeneutical marginalisation. The implication of cultural domination is that women are prohibited from exercising a core feature of their humanity, and as such are alienated from their species being. This account therefore posits that the wrongfulness in the gendered distribution of emotion work resides in the fact that it contributes to the oppression of women. The advantage of this account is that it situates the discussion of the unequal

73 Stellenbosch University https://scholar.sun.ac.za distribution of labour in a patriarchal reality, and thereby highlights the social significance of the problem, while at the same time accounting for the fact that women and men participate in this phenomenon largely unconsciously. It further does not depend on an account of exploitation that maps well onto the home or on an account of fairness. However, I would still like to explore the possibility of producing an account that could explain the wrongfulness of an unequal division of nurturing work even if the one doing more work is male, or, for example, if they are in a homosexual relationship. Further, if it is the case that the real problem is hermeneutical marginalisation as an aspect of status degradation, then we ought to be seeking the most efficacious interventions into that problem. It is probably the case that exclusion from positions of public power is more important than an unequal status accorded in the home. If so, then interventions such as affirmative action, particularly in fields that focus on knowledge-creation such as academia, or supporting women-run media forums such as movies, television, and literature would be more effective in allowing the participation of women in the production of social meanings. If that is the case, and these actions would constitute a sufficient intervention into hermeneutical marginalisation, then that would justify leaving the status quo of gendered division of nurturing work as is, except insofar as these other interventions may indirectly affect it.

CONCLUSION

This chapter gave an account of two kinds of exploitation that women face due to the unequal distribution of nurturing work in the home: an economic account of how women are exploited under capitalism via their work in the home, and a moralised philosophical account of exploitation that shows women are exploited because the unequal distribution of nurturing work degrades them. In exploring the economic account of exploitation, I also looked at the particular kinds of alienation that arise when what is being exploited is one’s emotional labour, and extended this to the home. Though there are possibly some circumstances in the home which could replicate the alienation that occurs in the workplace, those circumstances would only occur for a minority of women, and further likely present for only a short period of those women’s lives. In short, it can only account for the wrongfulness of such

74 Stellenbosch University https://scholar.sun.ac.za exploitation if a woman is so overworked that she has no resources to invest sincerity into her emotion work, whereas we need an account that will explain the wrongfulness of a far less drastic scenario. I then moved away from the kinds of exploitation arguments that are limited to the workplace to look at some other accounts of exploitation and how they explain the wrongfulness of exploitation. Wertheimer’s account of exploitation proved unfruitful because it is a conventionalist account where the status quo is used to determine fairness. Since we are working with a case where the status quo is what is unfair, given the unequal ‘relationship market opportunities’ of men and women, the wrongfulness of playing according to a (unfair) market cannot be explained. Goodin’s account is similar to Wertheimers’s in that it relies on convention to determine whether exploitation takes place, but with the caveat that it may sometimes be necessary to suspend convention in an engagement in the event of vulnerability. I showed that since women in these kinds of romantic relationships are not especially vulnerable, this caveat does not apply, and so we are left only with Goodin’s conventionalist account, which is no better than Wertheimer’s. We are then left with our two most promising accounts, which do account for some of the wrongfulness of unequal nurturing work. Sample’s account defines exploitation as degradation, where someone’s intrinsic work is not respected in an interaction. One of the ways in which this happens is if someone benefits by taking advantage of an existing injustice. This shows the wrongfulness of a gendered distribution of nurturing work in a patriarchal society, and imposes a strong moral obligation to refrain from exploiting someone in this manner. The wrongfulness is dependent on the existing injustice, and, as such, one cannot conclude from this that an unequal distribution of nurturing work is wrongful as such. Finally, the last argument suggested that an unequal distribution of nurturing work is wrong insofar as it contributes to the oppression of women via asymmetrical status accord and hermeneutical marginalisation. Though I found this argument convincing, it still seems to be missing something in the wrongfulness of such exploitative distribution of labour, since it implies that when a man exploits his female partner, he only wrongs her insofar as she is a member of a group who is oppressed by hermeneutical marginalisation. It seems to me that it should be possible to produce an account that explains a way in which a woman as an individual can be wronged by her partner exploiting her by not reciprocating her emotional work.

75 Stellenbosch University https://scholar.sun.ac.za

Secondly, once again, this can only explain the wrongfulness of men exploiting women, and not the wrongfulness of an unequal distribution of nurturing work in general.

76 Stellenbosch University https://scholar.sun.ac.za

CHAPTER FOUR: THE ETHICS OF CARE

INTRODUCTION

The previous chapter showed that a gendered division of nurturing work is wrong for two reasons: firstly, it is exploitative because men benefit from an existing injustice, and secondly, because it contributes to the oppression of women by reducing their status. This chapter consists of the final route that will explore the wrongfulness of an unequal distribution of nurturing work to ascertain if it is wrong as such, or only when it is gendered, and it does so by looking more closely at the nature of nurturing work. I will begin by attempting to ascertain whether or not any distribution of nurturing work within the heterosexual couple could be acceptable, or whether such a distribution needs to be more or less equal. I use Michael Sandel’s arguments in his book, What Money Can’t Buy (2012) to argue that financial disparity in couples is, at the very least, insufficient to justify an unequal division of nurturing work, and also show that nurturing work is not work with mere utility that could be performed by someone paid to do so. I still need to establish whether an unequal division of nurturing work is wrong per se. In section 4.2, I turn to the ethics of care as formulated by Joan Tronto in order to determine what moral obligations can be extracted from it. This chapter will look at the insights that the ethics of care gives particularly about how the peculiar nature of care gives rise to moral discourses and obligations. In order to make this useful for my discussion, I will need to show that there is sufficient overlap or similarities between what I have dubbed nurturing work and care. I believe this can easily be done, and so intend to apply the ethics of care to nurturing work. If I can show that there is a general obligation to reciprocate care, and if I can extend this to nurturing work, it would account for the wrongfulness of an unequal division of nurturing work.

4.1. MONEY CAN’T BUY YOU NURTURING WORK

Care in the context of intimate relationships seems to be the kind of thing that cannot be compensated for with money. In other words, it seems wrong that one partner does all the caring and is not taken care of, even if she earns much less than her husband

77 Stellenbosch University https://scholar.sun.ac.za and is therefore indirectly financially compensated for her care work. This point is not uncontroversial, so I will argue that this is the case below. I have chosen two kinds of arguments to show that this is the case. There are doubtless some more arguments, but I will argue this using, firstly, Michael Sandel’s arguments about why some things cannot or ought not to be monetised. Secondly, I will look at the account given about the nature of care by Joan Tronto, and show that because care is a particular kind of moral activity it cannot be outsourced. In his book, What Money Can’t Buy (2012), Michael Sandel argues that some things cannot or ought not to be monetised for two main reasons: firstly, the monetisation of that thing could lead to coercive practices given unequal power in a market, and secondly, because the introduction of money can degrade it, or crowd out the morality associated with it (110). Both are applicable here, and we will begin with how the monetisation of nurturing work can lead to coercion. One example of how monetisation can lead to coercion is related to the sale of organs like kidneys – if a kidney can be purchased, it may well improve allocative efficiency, and furthermore be beneficial to those who donate the kidneys. However, if this is the case, then those whose kidneys will be bought are likely to be the very vulnerable. Most people would not choose to make money by undergoing painful surgery, and if people do, it will probably be the result of coercion. Nurturing work is not directly analogous, since doing nurturing work is by no means as painful or as risky as donating kidneys. However, the monetisation of nurturing work can be coercive. If it is possible to ‘buy’ oneself out of nurturing work since one earns more money, many men will be able to do so. In most couples men earn more than women, and this is the consequence of existing injustice: firstly, there is the gender pay gap, where women earn significantly less than men, even in the most gender-equal countries. Secondly, we saw in chapter two that women are made vulnerable in anticipation of marriage. In terms of economics, people in non-coercive, already fair market conditions will make choices to maximise their utility. Under these conditions, then, women would agree to do extra nurturing work for money because it is worth the money for them to do so. However, this description of these arrangements is incorrect: rather than doing care work for the money, women are ‘obligated’ to do more care work because their husbands earn more than they do. Therefore, because of her inferior financial position, a woman is coerced into doing more nurturing work. We saw in

78 Stellenbosch University https://scholar.sun.ac.za chapter 1 that even women who earn more may do more nurturing work than their partners in order to preserve her partner’s dominance (‘power balancing’). It may be the case that some women would prefer to earn more doing other kinds of work, but perhaps their earnings are limited because they take on more responsibility for housework or childcare, or perhaps because they chose dead-end jobs or low-earning careers because of their gendered expectations of marriage. If the woman’s lower earnings is not even due to her different number of paid work hours, or different kind of career, but rather due to the gender pay gap, then it is especially clear that women should not have to do extra nurturing work to ‘pay off’ their husbands for their higher earnings. The second reason that Sandel gives for why market norms are not always appropriate is because the introduction of market norms can degrade the activity by putting a price on it. Some of the examples he mentions is the ‘purchasing’ of a baby, where the most desirable child goes to the highest bidder, rather than working through the normal process of an adoption agency (2012:111). This is wrong, he argues, because putting a price on an infant degrades the infant, and perhaps also the opportunity to be a parent. If it is true that ‘paying’ for nurturing work in a relationship corrupts it, it means that the introduction of market norms degrades nurturing work in some way. Therefore, if the wealthier partner ‘buys’ their way out of nurturing work, they are effectively labeling nurturing work as something that can be bought, and perhaps treating it as such degrades it. This may be an empirical question: do people value nurturing work less if they have paid for it? Is nurturing work less effective because it has been paid for? Of course, nurturing work has already been commodified to a large extent. We have seen that the friendliness of air hostesses constitute emotional labour. Care is commodified in the form of old age homes, nurses, child daycare or nannies. However, the question is not whether or not nurturing work can be commodified as such, but whether or not nurturing work in a relationship can be commodified, without some kind of loss. There are a few reasons why we might think that paying for it can corrupt nurturing work, and two of those are the ways in which care is analogous to some other examples Sandel raises of degradation by payment. Firstly, Sandel explores the phenomenon of gifts: gifts are puzzling because they do not cohere with the notion of allocative efficiency – an individual is almost always better at knowing what they

79 Stellenbosch University https://scholar.sun.ac.za want than their friends are. In fact, it seems that the receivers of gifts would pay less for gifts they receive than the giver did (in other words, they underestimate the cost of the gift) (100). For this reason, from an economics perspective, it is more efficient to give people money instead of gifts so that they can purchase the things that they want the most. However, giving money is qualitatively different to giving gifts. This might be because gifts are not merely transferring objects, but rather, the gift does extra work on top of the monetary value of the object: it serves to perform love, and to cement a relationship. Further, if one were to repay someone for a gift, that would change the nature of the gift: instead of representing all of these benefits for the relationship, it would turn it into a simple transaction, which effectively cancels the gift. A further (and probably related) example is that of sex. Sex can be monetised in the form of sex work, but it is once again qualitatively different to non-paid sex which is given freely, even enthusiastically, which affirms the value and status of the receiver. One need not rely on a significance view of sexuality9 to claim that sex means different things in these different contexts, and if sex within a relationship were paid for like sex work, this would count as degradation. Nurturing work may be analogous to these two things: gifts and sex both mean particular things in a relationship, and do work that contributes to the quality of the relationship. Replacing gifts or sex with transactions removes all the work that these things can do for a relationship. In a similar way, it is plausible that if nurturing work done in a relationship becomes a transaction, it will no longer do the extra work for the relationship, such as cultivating intimacy, letting someone feel loved, and demonstrate love and affection. If the caregiver does care work because she is being paid for it, and if the care-receiver feels entitled to care work because he is paying for it, it can no longer perform those functions for a relationship. Further, one cannot buy oneself out of that work because the goods offered by nurturing work cannot be reduced to mere utility that can be allocated elsewhere. In other words, it seems that nurturing work is good in and of itself, which cannot be substituted for other kinds of goods, and also cannot be bestowed by a surrogate. Nurturing work cannot be reduced to mere utility, and must be done by those party to the relationship, and as such is non-distributable.

5 The significance view claims that sex is only morally acceptable if it is an expression of romantic love (Benatar, 2002:192).

80 Stellenbosch University https://scholar.sun.ac.za

There is further reason to think that the monetisation of nurturing work will involve a serious cost, and that is the moral nature of nurturing work. This is a claim made by the ethics of care – that care has a particular moral value, and this claim can be extended to nurturing work. Sandel argues that in some cases, market principles crowd out morals, and so the moral value of nurturing work can be crowded out by market principles. For example, in Switzerland, a nuclear waste site needed to be found, and when the best site was found, consent had to be acquired from the people living close to the site (Sandel, 2012:115). A slight majority consented to the presence of the nuclear waste site. However, when this consent was incentivised – in other words, people would benefit financially from the proximity of the site – the number of those consenting dropped. Sandel argues that this is because, where there is no money involved, consent was seen as a civic sacrifice done for the good of the community, whereas when financial incentive was introduced, it seemed that consenting would mean taking a bribe to accept the degradation of the community (117). This could be applied to care in two ways: firstly, if the caregiver is paid for her nurturing work, then it is less likely that she will see her work as achieving a moral value. It seems likely that women who do more nurturing work in their families would reject any claim they do it because of the money, and it would degrade the work in their eyes: they do it because they believe it has moral value in their relationships. Of course, we know that women do more nurturing work because of gendered expectations, and the lower status of women likely also means they need to do more nurturing work. However, at least at a psychological level, many women probably do more nurturing work because they want to do something good for the people they love, and because they are interested in building and maintaining relationships, which is considered a moral activity by the ethics of care. In other words, they have a personal investment in doing the work, which far exceeds the monetary worth it might also entail. Finally, emotion work is irreducible and impossible to outsource. Cooking and cleaning is work that can be outsourced if one has reasonable means, or which can be relieved by technology, such as a washing machine or a dishwasher, or ready-to-eat meals and so on. Even grocery shopping can be automated. These kinds of work can be simplified, and take up less and less time. On the other hand, emotion work seems to be irreducible: every attempt to streamline, outsource, or automate it will also degrade it. Lessening the effort and work that goes into the activity is not desirable in the same way that doing so for housework would be. Rather, the effort or labour put

81 Stellenbosch University https://scholar.sun.ac.za into a relationship demonstrates the importance of the other person and of that relationship. This is significant because it means that the problem of division of nurturing work is inherent to it, and so cannot dissolve like the division of housework, which will become less and less if better technology becomes available and more affordable. Further, it is not clear that emotion work is the kind of thing that one can pay other people (who are not in the relationship) to do – one cannot outsource the cultivation of friendships, or the responsibility for someone’s health (at least beyond the domestic unit), or emotion work in a relationship. There are of course examples of emotion work that are monetised, such as therapy, customer service and its concomitant status elevation, or nursing, but even these cannot make one feel loved or sufficiently supported in the relevant way – we cannot be affirmed as the individuals that we are in these kinds of work. Monetised care work such as therapy and nursing are usually reserved for extreme circumstances, and are not necessarily equipped to anticipate needs in the same way that an intimate partner can do. Finally, and most obviously, the kind of work that care does to cement a relationship or cultivate intimacy can only be done by those party to that relationship.

4.2. THE ETHICS OF CARE

4.2.1 THE NATURE OF CARE

I have tried to show throughout this paper that nurturing work is in fact work, but I have also argued that nurturing work that occurs in a relationship is qualitatively different to other kinds of work. I have done so by pointing out how it is not like commercialised work. It is time now to turn to a more positive account of what is particular about care. Writers in the ethics of care have done much work to identify those things, and I will explore that now. I will use Joan Tronto’s formulation in her book, Moral Boundaries (1993), for my understanding of care ethics. In order to use the ethics of care to provide insights into the moral nature of nurturing work, I need to show that the similarities between how care-ethicists define care and how I define nurturing work are sufficient to extrapolate those insights to nurturing work. It is difficult to find one standard definition of care – care is defined by different theorists as an attitude, a virtue, a value, a kind of work, a practice, an

82 Stellenbosch University https://scholar.sun.ac.za activity, and more (Held, 2006:32-33). Tronto defines care as a ‘species activity that includes everything that we do to maintain, continue, and repair our ‘world’ so that we can live in it as well as possible’ (1993:103). This definition is so broad that it would be difficult for it not to include nurturing work. However, Tronto’s other writing on care does not suggest that she thinks that care is such a broad activity. Firstly, she claims that care seems to ‘involve taking the concerns and needs of the other as the basis for action’ (105). Secondly, she shows that caring has four elements or phases that need to be present for it to constitute the relevant moral activity, and the presence of these four phases seems to narrow the kinds of activities that constitute care beyond the act of building or maintaining the world. The four phases of caring correspond to four ethical components. The four phases are as follows: i).‘caring about’, which entails noticing the need to care; ii) ‘taking care of’, which entails taking responsibility for care; iii) care-giving, which is the actual work of care, and iv) care-receiving, which is being attentive to the response of the care-receiver in order to determine if their needs have in fact been met (106-108). There are four moral elements to caring, each of which is required by a phase of caring. Firstly, there is that of attentiveness, which corresponds to ‘caring about’: this involves being skilled at noticing when care is required, or alternatively stated, noticing needs (1993:106). Since being attentive is a moral activity, a failure to notice a need is a moral evil, and conversely, becoming good at noticing needs is a moral achievement (127). The second moral element is that of responsibility, which of course corresponds to ‘taking care of’. Tronto claims that responsibility is different to an obligation, since it is not determined by moral duties, but is rather context-dependent – the context will determine if someone feels responsible for care work and therefore take responsibility for it. For instance, if one notices a need that no one else will meet, this may or may not result in one feeling responsible for meeting that need (132). Responsibility is rooted in ‘political motivations, cultural practices, and individual psychology’ (132). Taking responsibility for someone’s need can also involve delegating the actual of work of doing so to the appropriate agents. For example, one could remind one’s spouse to exercise or make an appointment with a doctor. The third moral element is competence, where one’s caregiving must be successful at making the care-receiver feel taken care of. This means that the intention to give good care or even an attempt to care is insufficient: one must also be

83 Stellenbosch University https://scholar.sun.ac.za competent at caregiving. There is thus an obligation to become good at caring. The final element is responsiveness, where the carer must listen to the response of the care-receiver, must ‘consider the other’s position as the other expresses it’ (132), and amend the care given in response to what the other expresses. To summarise, for appropriate caring, the carer must identify needs accurately, take responsibility for those needs, be successful in alleviating those needs, and listen to the care-receiver if they express different needs or preferred ways of those needs being met. Tronto asserts that in order to provide good care, all four of these elements need to be present, providing care which is integrated (255). I would argue that nurturing work certainly possesses all of these phases and elements of caring. Each of the activities (housework, emotion work, and what I have referred to as care) entails noticing needs, whether that need is to repurchase an item that is running out in the kitchen, or putting in a load of laundry because someone is on their second-last shirt, making sure someone eats enough protein or takes their medication, or noticing that one’s partner is unhappy and worn out. Secondly, beyond noticing the need, it entails making the need your problem – taking responsibility for addressing the need. Thirdly, the actual doing of the nurturing work involves competence – the work is not done if the need was not successfully met. Finally, nurturing work must also entail responsiveness, in the sense that it makes no sense to express care towards one’s partner in a way that is not received as caring. For example, someone notices her partner is stressed out and upset, and knows that she likes to unwind by talking about her concerns over a glass of wine, and so pours her partner a glass of wine and asks him what’s wrong. If in fact what he would require in order to feel less stressed out is to be relieved of his responsibilities for an hour so that he can exercise, take a bath, or read a book, then her attempt to care for his needs would be ineffective. I would suggest that, based on this description of care, the nature of nurturing work is sufficiently similar to care to apply the moral insights from care to nurturing work. At this point I would like to remind the reader that I am trying to establish the wrongfulness of an unequal distribution of nurturing work. My discussion of the distinctive features is an attempt to determine if there is a moral obligation that arises merely from the nature of care itself, which would then make it a moral failure to fail to meet this moral obligation, and hence explain the wrongfulness of doing less nurturing work than your partner. Since the ethics of care specifically defines care as

84 Stellenbosch University https://scholar.sun.ac.za a particular kind of moral activity, I will attempt to determine what kinds of moral claims can be made based on this theory. The ethics of care has one characteristic that makes it unable to impose strong, universal moral obligations: its emphasis on context and existing relationships. As Tronto puts it, ‘we are better served by focusing on a flexible notion of responsibility than we are by continuing to use obligation as the basis for understanding what people should do for each other’ (133). Recall that the second element of care is that of responsibility, and Tronto specifies that responsibility is rooted in context, and will be determined by contingent factors, including culture and gender roles. This claim is descriptive rather than normative: Tronto does not argue that anyone ought to take responsibility for anyone else, but rather that some people might feel responsible and therefore choose to take responsibility. This approach is inherently conservative, since it uses the status quo as its starting point largely uncritically. Further, if universal moral values are eschewed in favour of context, there are no grounds on which the context as it is can be critiqued. A further example of this phenomenon in ethics of care can be seen in the work of Eva Feder Kittay, who tries to derive strong moral entitlements from the ethics of care. I will look at her chapter ‘Vulnerability and the Moral Nature of Dependency’ (2005) to show how she uses the ethics of care to determine what moral claims can be made about dependency work. Briefly, dependency work is the care work done for the vulnerable, such as the very young, the very old, or the infirm. Hers is chiefly a political theory which relies on the entitlements of citizens, and thus, unlike Tronto, she makes use of universal moral principles. Nevertheless, it will be shown in the next paragraph that her theory is equally conservative due to its similar reliance on context. Kittay asserts two kinds of entitlements: the first is that every person is entitled to the kind of relationship in which they may be cared for, and secondly that everyone is entitled to be supported in their role as carers (2005:273). She states that she is less interested in how people come to acquire an obligation to care, and rather starts from those who are already in relationships where they care for others, and looks at how they are to be supported in that. The reason for this is that caring is essential, but sometimes very costly to the care-provider. In order to prevent these costs from becoming unreasonably taxing, adequate support must be provided for the care-provider.

85 Stellenbosch University https://scholar.sun.ac.za

Since Kittay is interested in dependency work, the care-receivers in this kind of work are usually incapable of reciprocating care, and so cannot themselves provide the support their carers need (266). As such, standard reciprocity where the other party responds in kind, in the same amount and immediately afterwards, cannot be applied here. Instead, she suggests a model of reciprocity-in-connection, which is a nested model of care (273). The idea is that every caregiver must also have a caretaker. She does not suggest outsourcing this to government, but rather thinks communities must be organised in this way, and she believes they will be organised in this way because some people already feel responsible to care for others. For instance, one woman says that she has to take care of her elderly mother because her mother cared for her mother, and there is no one else to do it, so because her mother played the role of daughter caring for mother, the woman in question also has to play the role of daughter caring for her mother (273). This is reminiscent of Tronto’s notion of responsibility: this woman feels responsible for the care of her mother, but not because she has a moral obligation to do so. In other words, Kittay does not argue that one has an obligation to care for one’s parents, or for other persons generally, but rather argues that if one already feels responsible for the needs of another, then one will take responsibility for doing so. It is thus a factual question, ‘who feels responsible for whom?’, rather than a normative question, ‘who has an obligation to care for whom?’. It should be clear at this point that the fact that this woman feels responsible to care for her mother is a case of gendered injustice, since men in general do not seem to feel similarly responsible. The idea of reciprocity-in-connection instead of reciprocity means that as long as the carers are cared for, the fair allocation of care is superfluous. Since we know that women do more caring than men, female carers are likely to be taken care of by other female carers, while men are also taken care of by female carers, without being nested themselves in the responsibility to care for specific others. Further, it is hard to locate responsibility to provide care: Kittay mentions that if an existing relationship of care has moral warrant, then the carer has an obligation to care (272). It is unclear what kind of prior relation would have moral warrant. Beyond that, she claims that we are entitled to care and support for our caregiving, but she does not assert a corresponding duty for the entitlement. It is therefore not clear that if someone is entitled to the kind of relationship where they can be cared for, that any individual person has a corresponding duty to provide that

86 Stellenbosch University https://scholar.sun.ac.za care. Caregivers are also entitled to support for their care, but it does not seem that any duty is placed on individuals to provide that care. As such, the entitlement cannot do any work. Let us take an example of someone who has a mental illness such as depression. It seems right to say that if such a person is so depressed that she cannot reciprocate care, this should not exclude her from receiving care. One could perhaps even go so far as to say that such a person is entitled to be cared for. However, this surely does not imply that when a stranger encounters such a person, he is obligated as an individual to befriend her and to provide her with care. We could make a more general statement that quality mental illness institutions such as psychiatric institutions must be made available for those who need it, which would impose a corresponding duty on the state, and consequently the entitlement to relationships where care is typically provided is empty. Further, saying that someone is entitled to be cared for when they are suffering from mental illness (which could be addressed with psychiatric institutions) is different to saying that someone is entitled to the kind of relationships where care is provided. Of course, families often do fulfill this role, but it is strange to say that those without families are entitled to other non-relatives to fulfill the role that family would beyond childhood. Kittay’s work is thus a good example of the kinds of obligations that can be established by an ethic of care. It makes sense that Kittay argues for a different model of reciprocity, given that her focus is dependency work, but since I am interested in men and women who are capable of caring and yet distribute nurturing work in a gendered way, I need a more robust form of reciprocity that can impose an obligation on a specific individual towards another specific individual. Let us then turn back to Tronto’s understanding of ethics of care. She claims that the ethics of care is not interested in the question ‘What, if anything, do I (we) owe to others?’, but rather the question, ‘How can I (we) best meet my (our) caring responsibilities?’(137). Again, the important difference here is that ‘caring responsibilities’ is a factual question of who already feels responsible for meeting the needs of others, whereas the question of what we owe to others would establish a moral obligation to care for others, even if one does not already feel responsible for another’s needs. This means that she is explicitly not interested in determining when one has obligation to provide care, but rather starts with existing caring relationships and the responsibilities that seem to flow from them. Therefore, if one does not

87 Stellenbosch University https://scholar.sun.ac.za already believe that one has a responsibility to care, then there is no way to establish such a responsibility from the point of view of this theory. If this is the case, it will naturally be very difficult to establish an obligation to care. There is a greater problem, however: if it is a factual question of who already bears caring responsibilities, then the ethics of care only applies to women and the few men who are already carers. Tronto specifies that our caring responsibilities are determined by who we are and where we are situated. This is reminiscent of Kittay’s discussion on Goodin: she states that we should not use Goodin’s ‘pragmatic ought’, which states that existing relationships of responsibility should be considered a ‘social fact’, and moral obligations attach to those already in such relationships in the event of the other’s vulnerability (Goodin in Kittay, 269). Kittay discards the notion of the ‘pragmatic ought’ precisely because social facts are so often oppressive. Kittay nevertheless roots the responsibility to do dependency work in the following factors: 1) basic needs need to be met, 2) a person is very vulnerable, and 3) ‘the prior relation, which puts a particular individual into position of having to assume such a vulnerability-responsive obligation, has a moral warrant’ (2005:272). At no point in Kittay’s chapter does she explain beyond the fact of vulnerability how a dependency relationship has moral warrant. She seems to include this provision to ensure that slaves cannot be beholden in this dependency- relationship with slave-owners. Perhaps it is possible to argue that, given that the greater burden of care on women is an injustice, their care-responsibilities have no moral warrant. However, since many of these are still intimate relationships, it simply seems untrue that the dependency relationship has no moral warrant. Regardless, Kittay does not make these claims about the responsibility to care generally, but only about dependency work, whereas Tronto makes no attempt to show how people have caring responsibilities beyond the social fact of their gendered burden. The ethics of care as formulated by these two theorists is thus plagued by a problem: it is vague in its allocation of responsibilities and cannot convincingly situate an obligation to care. Because of this, despite being a feminist ethic, it is not useful in arguing for a fairer and less gendered allocation of nurturing work.

4.2.2. CARE AS A VIRTUE

88 Stellenbosch University https://scholar.sun.ac.za

In order to address the shortcoming of vagueness in terms of moral responsibility, Raja Halwani attempts to subsume care into virtue ethics and define it as a primary virtue, which means that it is a virtue ranked more important than many other virtues, and is essential for living well (2003:183). She argues that in order for care to be considered a virtue it needs to adhere to the definition of a virtue and bear the criterion of a virtue, which is that it must contribute to human flourishing (182). Aristotle’s definition of virtue is a ‘state involving choice and lying in a mean, with a mean relative to the individual’ (182). In other words, care needs to be the kind of thing that one can choose to do ‘at the right times, about the right things, toward the right people, and in the right way’ (182). Care certainly appears to match this: it should be clear by now that care is a choice, rather than a personality trait that one either has or does not have. Care can fall into a mean: it is possible to provide too much care and be smothering, and it is possible to provide too little – in other words, care can be defective or vicious (182). There are appropriate times and circumstances to offer care, and the right and wrong people to offer care to. Halwani offers a modern definition of virtue as well, as conceptualised by Linda Zabzebski: virtue is ‘a deep and enduring acquired excellence with a characteristic motivation to bring about a certain desired end and reliable success in bringing about that end’ (Zabzebski quoted in Halwani, 2003:182). Care is certainly this kind of thing as well: one can be good or bad at care, and become better at it, and it is only good care if it actually successfully meets someone’s needs. Secondly, care must satisfy the criterion of being a virtue: it must be linked to eudaimonia, where, along with other virtues, care is causally effective in bringing about human flourishing for the moral agent (183). Human relationships are essential in bringing about flourishing, and care is essential for human relationships, and so it assists in human flourishing. Further, if we take the insights of the ethics of care seriously, where offering care and being cared for are essential parts of the human experience, then caring is also a way of being characteristically human. It thus seems plausible that care be subsumed under virtue ethics without too much lost, and much clarity gained. There is much debate on this subject, but it seems to me that care ethics as a kind of virtue ethics is the strongest formulation of care ethics, since it escapes much of its original vagueness. It further manages to elevate the status of care as an important moral activity that forms part of the right way to live.

89 Stellenbosch University https://scholar.sun.ac.za

However, even such a strong formulation of the ethics of care does not manage to solve the problem of addressing an unequal distribution of such work. Firstly, if care is a virtue, women get some benefit from doing more of it – they become more virtuous, and they would tend to flourish more and experience more happiness and fulfillment. If this the case, it is not clear why a woman should complain about doing more care work than her partner. If she is both improving her moral character as well as increasing her chances of human flourishing, a woman should be happy to do care work even if her partner does none. We can compare this to Jean Hampton’s contractarian test from Chapter 2: testing for exploitation by determining whether the costs and benefits are distributed in such a way that a self- interested person can accept it is contrary to the ethic of care. This is because a woman is being given an opportunity to become virtuous, which is a good apart from whether or not her relationship is fair to her, or if she is taken care of. In fact, if a woman does see it this way, this will contribute to her reluctance to insist on being cared for. Though the conceptualisation of care as a virtue elevates the status of the care work that a person does, it seems that the consequence of this is the valorisation of women who engage in care work, which reinforces the ‘angel in the household’ trope. This trope refers to the kind of self-sacrificing woman who cares for everyone without being cared for in return. It therefore makes it seem that she ought to be sufficiently praised for her moral goodness, but it certainly does not impose an obligation on others to care equally well for her. However, insofar as care is a virtue, it perhaps imposes a weak obligation to be a good carer on men and women alike. In order to be a virtuous person, one must be a good carer. This formulation is different from a deontological formulation, which would frame this as follows: one has a duty to refrain from exploiting others, and so one must be a good carer; or, by not reciprocating one’s partner’s care, one is wronging one’s partner by failing to discharge one’s duty to them. The virtue ethics formulation also does not allow us to look at the quality of the relationship, and whether or not the relationship flourishes. Rather, being a good carer and engaging in care work are important for the cultivation of a good moral character and in human flourishing. Let us then cash out this weak moral obligation.

90 Stellenbosch University https://scholar.sun.ac.za

4.3. A WEAK OBLIGATION TO PERFORM NURTURING WORK

We have thus established the following weak obligations: ‘in order to be a good person, one must be a good carer’, and ‘in order to flourish as a human being, one must be a good carer’. This naturally applies to men as well as women, and this means that men who are not currently good carers need to become such in order to be good people and in order to flourish as human beings. This weak obligation is still different to the notion of ‘responsibility’ offered by Kittay and Tronto, because it exists apart from the fact of being nested in relationships and already feeling responsible for others. In other words, it is not a factual claim, but a normative one. Given that, let us tease out what this means for unequal caring relationships which are unequal because one partner provides insufficient care work. Firstly, in order to cultivate the virtue of care, a number of component skills need to be acquired, and these are the four moral elements of care that Tronto identifies. A person needs to become good at attentiveness, taking responsibility for others’ needs, for effectively meeting those needs, and being responsive to how the care-receiver has received the care. This is useful because it refutes a common excuse for doing less nurturing work: that the person simply didn’t notice the other’s need, or tried but failed to meet that need. The fact that a man has not been socialised into being good at nurturing work does not alleviate the obligation to acquire such a skill. Further, by associating care with an individual’s virtue and therefore eudaimonia, it means that it cannot be a casualty to a pragmatic distribution of work. I am referring to the conversations that couples have justifying the current, unequal distribution of work, where the female partner would rather do the work herself because she is better at anticipating needs. For example, one woman I know takes on all the cooking, because even though her partner is a very competent cook, he is very bad at having food ready at a reasonable time when she is hungry, instead of, for instance, at 11 o’clock in the evening. While it is impossible to assert from virtue ethics that he has a particular moral obligation to his partner to address her hunger on the basis of care ethics, it would be necessary for the cultivation of his own virtue of caring to be more effective in meeting her needs. Subsuming being a good nurturer under individual virtue has a further advantage, which is that it means the responsibility to become better at nurturing work resides with the individual, and not his partner. Because of the way in which

91 Stellenbosch University https://scholar.sun.ac.za boys and girls are socialised differently, it is usually the case that adult men simply are worse at nurturing work than their female partners. This can often mean that if a woman wants her partner to be better at nurturing work, she has to train him to be. However, if being good at nurturing work is necessary to be a virtuous person with a chance at flourishing as a human being, then he should want to acquire those skills independently of what his partner wants. It can be seen that even though the obligation is not a strong one, it nevertheless is not insignificant. Care as a virtue gives us a reason to think that individuals must become good at caring, for their own good, and the ethics of care provides insight into what kinds of skills individuals need in order to do so. Regardless of this, it still cannot address the fairness of the distribution of care, and the unfairness of a woman who does all the caring for and is never taken care of.

CONCLUSION

This chapter highlighted two significant aspects of nurturing work: firstly, that nurturing work is not the kind of thing that can be monetised because it is irreducible and impossible to outsource. Secondly, nurturing work is a moral activity, and I have argued that it is most usefully thought of as a virtue. Let us pause for a moment at the first aspect I have highlighted. Crucially, nurturing work does not have mere utility in completing tasks that need to be completed, but also does work on a relationship, and this means that only those party to the relationship can do this work, and that the quality of a relationship is in part determined by the nurturing work performed by those in the relationship. This answers a question that has come up throughout this thesis: does one’s partner have a monopoly on certain kinds of nurturing work one receives? The answer seems to be yes – even if one is taken care of by people outside of the relationship, the intimate relationship itself can only benefit from work done by oneself or by one’s partner. This coheres with the empirical data I presented in section 1.2.2, which stated that the level of satisfaction felt in a relationship is strongly correlated with the equality of emotion work performed, not the total emotion work performed. This also means that financial inequality between partners cannot justify an unequal division of emotional labour.

92 Stellenbosch University https://scholar.sun.ac.za

The second characteristic of nurturing work is its moral nature. I had hoped that this would yield a strong obligation to others to provide nurturing work. However, the formulations of ethics of care that I have looked at are rooted in context, and on existing responsibilities to care, which I have argued are too conservative to argue for a more egalitarian and less gendered distribution of nurturing work. In order to address this shortcoming, I used Raja Halwani’s argument that an ethic of care is best conceived of as a kind of virtue ethics, where care is a virtue. Viewing care as a virtue is primarily advantageous because it situates the (weak) obligation to care in the individual, so an individual who wants to become virtuous and flourish in a characteristically human way needs to become good at nurturing work. Despite this advantage, viewing care as a virtue still cannot impose obligations on individuals to others, and cannot explain why providing insufficient nurturing work for one’s partner is a way of wronging one’s partner. Instead, what this chapter has managed to argue is that, firstly, a relationship is enriched by the performance of nurturing work, and that individuals become more virtuous by doing nurturing work well. Therefore, I cannot conclude yet that an unequal division of nurturing work in a relationship is wrong as such. Nevertheless, the conclusion to this thesis will bring the question of gender in again, and solidify the arguments that a gendered division of nurturing work is wrongful because it violates a strong moral obligation.

93 Stellenbosch University https://scholar.sun.ac.za

CONCLUSION

Most women who are with men do not believe that their partners are sexist. In fact, I am explicitly interested in men who hold egalitarian ideologies, and as such, are indeed not sexist (by which I mean they do not hold inferiorising beliefs about women). As such, it can be easy to chalk up the unequal distribution of nurturing work in heterosexual relationships to factors other than gender inequality: for instance, to preference, to personality, or to innate ability. The couples I’ve encountered will often explain their unequal labour distribution in this way: ‘she’s just better at cooking than I am’, or ‘when I clean, I don’t do it to her standard, so it’s better if she just does it’, or, ‘eating dinner at a reasonable hour is more important to me than to him, so I just take responsibility for dinner’. They might also say, ‘my wife has an amazing ability to intuit what I’m feeling and provide empathy. I just don’t notice this kind of thing’. I hope that the work in the first chapter, and in fact in the subsequent chapters, has made clear that nurturing work is not a skill that women are better at because of their innate abilities. Rather, women are ‘better’ at it because they have been socialised into taking on this role, and because they are held accountable for performing it in a way that men are not. In other words, it is likelier to become part of their ‘personality’ because they need to be good at nurturing in order to be likeable, and as such are constantly motivated to do it in a way that men are not. One need only think of the study I cited in the first chapter that showed that men are equally good at empathy when they have financial motivation to do it. I turned to Jean Hampton’s solution because it is intended to counteract this tendency in the way women are socialised to be self-sacrificing. She proposes that one should deliberately invoke one’s self-interest to determine if one’s relationship is exploitative. She proposes that one set aside one’s feelings of love and duty which might cloud one’s judgement, and make a sober assessment of what one has sacrificed for one’s partner. Though I believe she has identified the problem very accurately, her solution is inadequate. I argued that the exclusion of feelings of love and duty in assessing one’s relationship would make one’s behaviour appear utterly irrational, since those are precisely the reasons why one makes sacrifices in a relationship, whether you are a man or a woman. I further argued in the fourth chapter that

94 Stellenbosch University https://scholar.sun.ac.za nurturing work does not merely serve to complete a task or provide utility, but also does work on the relationship – in other words, these are tasks that are done in love and that foster love, and as such are excluded by the exclusion of feelings of love and duty. Okin’s argument is decidedly in the realm of political philosophy, and as such its excellence lies in identifying the duties of the state to address the injustice of the gendered structure of marriage. She shows how women are made materially vulnerable in anticipation of marriage, during marriage, and upon dissolution of marriage, and makes excellent suggestions of policies that would mitigate this vulnerability. She further suggests that workplaces need to be structured in such a way that makes it possible to have a more equal division of nurturing work but also of sacrifice between husband and wife. I endorse these policy recommendations entirely. The distribution of labour is unjust when it contributes to the material vulnerability of women, and it would be unjust of the workplace to prevent couples from achieving an egalitarian distribution of labour. However, this does not go far enough. I want to suggest that an unequal distribution of labour is wrong even if it does not render women materially vulnerable. To this end, I turned to theories of exploitation to show the wrongfulness of a gendered distribution of nurturing work. The work done by materialist feminists shows how women’s doing unpaid reproductive labour is integral to the functioning of capitalism. Though Marx saw exploitation as an ill of the proletariat, materialist feminism shows that women insofar as they are wives are also exploited, and that they constitute a class whose exploitation takes its own particular form, namely that a woman’s labour power is appropriated by her husband, with whom she bears a material relationship of serfdom. Therefore, capitalism is as bad for wives as it is for workers. However, this functions mostly as a critique of capitalism rather than a critique of the way people set up their relationships and households. It constitutes a structural critique of material relations, and not a critique of how men treat their partners, or indeed a statement on how labour ought to be distributed in the home. I thus turned to other definitions of exploitation that could show the wrong that the one who exploits does to the one exploited. The most productive definition is that of Ruth Sample, where she defines exploitation as a form of degradation, and an instance of such degradation is taking advantage of an existing injustice. This is certainly the case in a gendered distribution of nurturing work. That this distribution

95 Stellenbosch University https://scholar.sun.ac.za of labour is a matter of justice was argued in the second chapter, but briefly, the virtually automatic allocation of labour based on a morally arbitrary factor such as gender, plus the fact that this has a great effect on women’s lives, makes it an issue of injustice. It is an injustice that women are socialised into being self-sacrificing and good at nurturing work, whereas men are not socialised into emotion work. Men benefit from doing less nurturing work than their partners: as a result of it, they live in more harmonious homes that are a pleasure to dwell in (without having contributed equally to making it so), they are healthier and more integrated socially, and they have their status elevated and their negative emotions soothed from receiving emotion work. Given that men benefit from the status quo, this constitutes gaining an advantage from an existing injustice. Thus, even if such a distribution caused no harm, and even though the relationship is mutually beneficial, is still constitutes exploitation because the advantage is gained through degradation. There is a further way in which an unequal distribution of nurturing work constitutes degradation, and that is because it is a case not just of unequal work, but unequal status accord. Emotion work in particular constitutes a kind of status accord, and by not reciprocating emotion work, a man performs for himself, for his partner, and those around him, that he is of higher status than his partner. By doing more emotion work than her partner, a woman affirms the same thing. This frequent affirmation of the greater importance assigned to men than women is a way of contributing to the lower status of women, and thereby degrades women. This gives a suggestion of the particular way in which a man wrongs his partner by not doing as much emotion work as she does – he degrades her by refusing to accord her status. This act is nevertheless not merely a wrong that occurs between two individuals. I argued that it contributes to a kind of structural injustice known as hermeneutical marginalisation. I used Bartky’s argument that emotion work involves an ‘epistemic lean’ that entails a woman ‘seeing the world according to him’, and this epistemic lean constitutes a displacement of a female perspective for a male one. If this is not reciprocated, then men continue to have an unfairly large role in generating cultural and social meanings. It is particularly significant in the generation of hermeneutical resources, which is precisely what is used to communicate ‘the world according to him’. For women’s perspectives to be allowed to help generate cultural meanings and hermeneutical resources, the ‘world according to her’ needs to be given as much status as the ‘world according to him’. The gendered unequal distribution of

96 Stellenbosch University https://scholar.sun.ac.za nurturing work therefore contributes to the hermeneutical marginalisation of women, and as such, constitutes a kind of injustice. These three kinds of degradation happening over and over in the microcosm of the family normalises the oppression of women by diminishing their status, denying them the advantage in participating in the generation of hermeneutical resources, and takes advantage of the already diminished status of women. I believe this is sufficient to show that a gendered distribution of nurturing work is wrongful, and that it wrongs both individual women and women as a group. Given the obligations to avoid exploiting others and contributing to the degradation of others, there is a strong moral obligation on men to refrain from allowing their relationships to display a gendered distribution of labour. A gendered distribution of nurturing work thus constitutes a moral harm committed by men against their female partners. It is clear that these strong moral obligations derive from the fact that there are existing background conditions of gender injustice, and moreover, a gendered distribution of labour normalises and contributes to the oppression of women, which increases the moral urgency to oppose it. I have thus answered part of my question: at least when an unequal distribution of labour is gendered, it is wrong, and therefore there is a moral obligation to refrain from exploiting and degrading one’s partner in this way. However, this does not tell us whether or not an unequal distribution of labour is wrong as such. If the genders were reversed, and the male partner did more nurturing work than his partner, it would not be wrong for the reasons listed above, since they are premised on the one doing more work doing so because of existing oppression. I thus attempted to use ethics of care to try and establish a general moral obligation to reciprocate nurturing work. I was unable to do this. The ethics of care prefers the notion of ‘responsibility’ to ‘obligation’, which I argued makes it conservative rather than transformative, since ‘responsibility’ is a descriptive claim about existing relationships and the psychology of the people involved in such relationships, rather than a normative claim. The ethics of care is thus extremely valuable in that it provides a nuanced account of what constitutes good care, but is unable to account for how caring relationships arise in the first place, and when they ought to arise if they have not already. I want to argue that an unequal division of nurturing work is wrong, and therefore the partner who does less ought to do more than (typically) he is presently doing. If the theory is

97 Stellenbosch University https://scholar.sun.ac.za constrained by the existing dynamics of relationships, it cannot transform such dynamics. It cannot prescribe the obligation to become a better nurturer, and as such dooms whomever does more nurturing work to continue doing more nurturing work. Nevertheless, this does not mean that the ethics of care provides no insight. If we conceptualise care as a virtue, and subsume it into virtue ethics, then there is a weak moral obligation of some sort to become better carers. I call this moral obligation weak because it is not something owed to others, but rather in service of becoming a virtuous person who is capable of flourishing. I agreed with Halwani’s argument that care bears the required characteristics of virtue: it is an activity as well as a disposition, one can become excellent at it through practice, and it is undeniably linked to the achievement of eudaimonia. As such, people ought to be good carers in the same way they ought to be honest, just, temperate, and generous. Being good at nurturing work constitutes part of the ‘the good life’ that will facilitate one’s flourishing. The seriously diminished welfare of widowers after the death of their wives discussed in the first chapter proves the lack of flourishing that is the result of not engaging in nurturing work throughout one’s life. Because virtue ethics takes as its subject the actor rather than the acts, one important dimension of ethics of care is lost, and that is the notion of care as a relational activity. We saw that nurturing work cannot be reduced to mere utility partly because it does work to strengthen the relationship, not just achieving an end such as having clean dishes, a new book, or flowers in a vase. Nurturing work allows not only individuals to flourish, but also improves the quality of the relationship, and as such contributes to the flourishing of the relationship. We saw in the first chapter that it is empirically the case that people experience more relationship satisfaction the more emotion work both partners do (and in equal amounts), which corroborates the idea that nurturing work serves to strengthen the relationship. There is surely no moral obligation to have good romantic relationships, but it contributes to the flourishing of a relationship, and therefore to the flourishing of those in the relationship. Where a man in a relationship with a woman does less nurturing work than his partner, he must do more of it in order to allow the relationship to flourish. If this entails acquiring skills or knowledge to do this, he has an obligation to do this. If he does so, he also becomes more virtuous and likelier to flourish, and moreover, the quality of the relationship will improve and his partner is more likely to flourish too.

98 Stellenbosch University https://scholar.sun.ac.za

Doing more nurturing work is not limited to the relationship of the couple: doing more nurturing work for his friends, neighbours, or extended family, as well as children, will also contribute to his own flourishing, and to the flourishing of those relationships. Let us consider again one distressing statistic from the first chapter: in the case of a surviving parent, if that surviving parent is male, they have little more contact with their children than they would have if they were deceased. In other words, they are nearly completely isolated from their grown children. If men contribute more nurturing work to the relationships with their children throughout their lives, they are likely to have better relationships with their adult children, to the likely benefit of both parties. The ethics of care gives us insight into what becoming better at nurturing work would necessitate: one would have to learn to be attentive, and notice the needs of others. One would have to learn to take responsibility for the meeting of those needs. One would need to learn to be effective in the meeting of those needs, and learn to be responsive to the feedback of the care-receiver. The work done thus far dispels completely the notion that being a good nurturer is a personality trait, with all its connotations of being ‘hard-wired’ or unchangeable. It dispels the notion that some people (particularly women) do more nurturing work because they like to do it – rather, they are expected to do it in a way that men are not. This thesis has shown that men can be expected to do as much nurturing work (quantitatively) as well as women do it (qualitatively), and in fact not to do so constitutes exploitation and degradation of their partners, and neglect of the quality of their relationships with them. As such, I have proven a moral obligation to refrain from entertaining a gendered division of nurturing work in one’s relationship. I have further elevated the status of nurturing work to a component of human flourishing, as well as the flourishing of relationships. Therefore, though there may not be a moral obligation to have an equal distribution of nurturing work regardless of gender, there is certainly a way in which not having that is bad, in that it deviates from what constitutes ‘the good life’, and inhibits the flourishing of one’s most intimate and sustained relationships.

99 Stellenbosch University https://scholar.sun.ac.za

REFERENCE LIST

Abbey, R. 2011. The Return of Feminist Liberalism. Acumen: Durham. August, K.J and Sorkin, D.H. 2010. 'Marital status and gender differneces in managing a chronic illness: The funciton of healht-related social control'. Social Science and Medicine, 71: 1831-1838. Bartky, S.L. 1990. Femininity and Domination. Routledge: New York. Berardo, FM. 1970. "Survivorship and social isolation: the case of the aged widower". The Family Coordinator, 19(1):11-25. Bittman, M, England, P, Sayer, L, Folbre, N and Matheson, G. 2003. 'When does gender trump money? Bargaining and time in household work'. American Journal of Sociology, 109(1): 186-214. Blackburn, S. 2005. The Oxford Dictionary of Philosophy. Oxford University Press: Oxford. Brook. P. 2009. 'The Alienated Heart: Hochschild's "emotional labour" thesis and the anticapitalist politics of alienation'. Class & Capital, 98: 7-31. Card, R. 2001. 'Gender, justice within the Family, and the Commitments of Rawlsian Liberalism'. Public Affairs Quarterly, 15(2):155-171. Colley, H. 2015. "Labour-power". In Marxism and Feminism, ed. Mojab, S. Zed Books: London. Chrisensen, A and Heavey, CL. 1990. "Gender and social structurei n the demand/withdraw pattern of marital conflict". Journal of Personality and Social Psychology, 59(2):71-81. Curran, MA, McDaniel, BT, Pollitt, AM, and Totenhagen, CJ. 2015. "Gender, emotion work, and relationship quality: a daily diary study". Sex Roles, (73): 157-173. Delphy, C. 1984. Close to Home: A Materialist Analysis of Women's Oppression. Hutchinson: London. Erickson, RJ. 2005. "Why emotion work matters: sex, gender, and the division of household labor". Journal of Marriage and Family, 67(2):337-351. Eriksson, R and Nermo, M. 2010. "Care for Sick Children as a Proxy for Gender Equality in the Family". Social Indicators Research, 97(3): 341-356. Fine, C. 2010. Delusions of Gender. Icon Books: New York. Fricker, M. 2007. Epistemic Injustice. Oxford University Press: Oxford. Galdas, P.M., Cheater, F. and Marshall, P., 2005. "Men and health-seeking behaviour: literature review". Journal of Advanced Nursing, 49(6):616-623. Goodin, R. 1987. 'Exploiting a situation, exploiting a person'. Modern

100 Stellenbosch University https://scholar.sun.ac.za

Theories of Exploitation, ed. Reeve, A. SAGE Publications: Bristol. Goodin, R.E. 1985. Protecting the Vulnerable. The University of Chicago Press: Chicago. Grall, T.S. 2009. Custodial Mothers and Fathers and Their Child Support: 2007. US Census Bureau: Washington, D.C. Halwani, R. 2003. 'Care Ethics and Virtue Ethics'. Hypatia, 18(3), 161-192. Hampton, J. 2007. The Intrinsic Worth of Persons. Cambridge University Press: Cambridge. Held, V. 2006. The Ethics of Care: Personal, Political, and Global. Oxford University Press: Oxford. Himmelweit, S. 1991. 'Reproduction and the materliast conception of history: A feminist critique', in The Cambridge Companion to Marx, ed. Carver, T. Cambridge University Press: Cambridge, pp. 196-221. Hochschild, A.R. 2003a. The Second Shift. Penguin Group: New York. Hochschild, A.R. 2003b.The Managed Heart. University of California Press: Berkeley. Hochschild, A.R., 1979. Emotion Work, Feeling Rules, and Social Structure. American Journal of Sociology, 85(3), pp.551–575. Kimmel, M. 2000. The Gendered Society.Oxford University Press: New York. Kittay, E.F. 2005. 'Vulnerability and the Moral Nature of Dependency Relations'. Feminist Theory, eds. Cudd, A and Andreasen, O. Blackwell Publishing: Oxford. Kurdek, L.A. 2005. 'What Do We Know about Gay and Lesbian Couples?'. Current Directions in Psychological Science, 14(5): pp.251-254. Mendus, S. 2000. Feminism and Emotion. Macmillan Press LTD: Hampshire. No Author. 2016. Statistics Sweden,Women and Men Sweden 2016. No Author. 2015. Emotional Labour: The Metafilter Thread Condensed. [Accessed on: 30/10/2017]. Available at: https://drive.google.com/file/d/0B0UUYL6kaNeBTDBRbkJkeUtabEk/view?pref=2& pli=1 O’Brien, R., Hunt, K. & Hart, G., 2005. “It”s caveman stuff, but that is to a certain extent how guys still operate’: Men’s accounts of masculinity and help seeking. Social Science and Medicine, 61(3), pp.503–516. Okin, S. M. 1989. Justice, Gender, and the Family. Basic Books, Inc: New York. Paoletti, I. 2002. "Caring for older people: a gendered practice". Discourse & Society, 13(6): 805-817.

101 Stellenbosch University https://scholar.sun.ac.za

Radzik, L. 2005. 'Justice in the family: A defence of feminist contractarianism'. Journal of Applied Philosophy, 22(1): 45-54. Sample, R. 2002. 'Why feminist contractarianism?' Journal of Social Philosophy, 33(2): 257-281. Sample, R. 2003. Exploitation: What It Is and Why It's Wrong. Rowman & Littlefield Publishers, Inc: Oxford. Sandberg, S. 2010. Why we have too few women leaders [Video File]. Retrieved from: https://www.ted.com/talks/sheryl_sandberg_why_we_have_too_few_women_leaders Sandel, M. 2012. What Money Can't Buy. Farrar, Straus and Giroux: New York. Thébaud, S., 2010. Masculinity, Bargaining, and Breadwinning: Understanding Men’s Housework in the Cultural Context of Paid Work. Gender & Society, 24(3), pp.330–354. Tichenor, V. 2005. "Maintaining men's dominance: negotiating identity and power when she earns more". Sex Roles, 53(3): 191-205. Tronto, J. 1993. Moral Boundaries. Routledge: London. Wertheimer, A. 1996. Exploitation. Princeton University Press: Princeton. Wiswall, M and Zafar, B. 2016. 'Preference for the workplace, human capital, and gender'. National Bureau of Economic Research Working Paper Series. World Economic Forum. 2016. The Global Gender Gap Report. World Economic Forum: Geneva.

102